Академический Документы
Профессиональный Документы
Культура Документы
Vo lu m e 3 4, N u m b e r 2 , 2 0 1 3
Contents
Articles
Data Science, Predictive Analytics, and Big Data: A Revolution That Will Transform Supply
Chain Design and Management
Matthew A. Waller and Stanley E. Fawcett
77
85
94
Measurement and Moderation: Finding the Boundary Conditions in Logistics and Supply
Chain Research
Thomas J. Goldsby, A. Michael Knemeyer, Jason W. Miller, and Carl Marcus Wallenburg
jbl_v34_i2_contents.indd 1
109
117
131
Managing the Innovation Adoption of Supply Chain FinanceEmpirical Evidence From Six
European Case Studies
David A. Wuttke, Constantin Blome, Kai Foerstl, and Michael Henke
148
167
6/3/2013 12:00:02 PM
The Publisher, The Council of Supply Chain Management Professionals, and Editors cannot be held responsible for errors or any consequences
arising from the use of information contained in this journal; the views and opinions expressed do not necessarily reect those of the Publisher The
Council of Supply Chain Management Professionals and Editors, neither does the publication of advertisements constitute any endorsement by the
Publisher, The Council of Supply Chain Management Professionals, and Editors of the products advertised.
Copyright and Photocopying
Copyright 2013 The Council of Supply Chain Management Professionals. All rights reserved. No part of this publication may be reproduced,
stored or transmitted in any form or by any means without the prior permission in writing from the copyright holder. Authorization to photocopy
items for internal and personal use is granted by the copyright holder for libraries and other users registered with their local Reproduction Rights
Organisation (RRO), e.g. Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923, USA (www.copyright.com), provided
the appropriate fee is paid directly to the RRO. This consent does not extend to other kinds of copying such as copying for general distribution,
for advertising or promotional purposes, for creating new collective works or for resale. Special requests should be addressed to: permissionsuk@
wiley.com
Delivery Terms and Legal Title
Where the subscription price includes print issues and delivery is to the recipients address, delivery terms are Delivered at Place (DAP); the recipient is responsible for paying any import duty or taxes. Title to all issues transfers FOB our shipping point, freight prepaid. We will endeavour to
full claims for missing or damaged copies within six months of publication, within our reasonable discretion and subject to availability.
Information for subscribers
Journal of Business Logistics is published quarterly. Institutional subscription prices for 2013 are:
Print & Online: US$ 281 (US), US$281 (Rest of World), 231 (Europe), 199 (UK). Prices are exclusive of tax. Asia-Pacic GST, Canadian GST
and European VAT will be applied at the appropriate rates. For more information on current tax rates, please go to www.wileyonlinelibrary.com/
tax-vat. The price includes online access to the current and all online back les to January 1st 2009, where available. For other pricing options,
including access information and terms and conditions, please visit www.wileyonlinelibrary.com/access.
Back issues: Single issues from current and recent volumes are available at the current single issue price from cs-journals@wiley.com. Earlier
issues may be obtained from Periodicals Service Company, 11 Main Street, Germantown, NY 12526, USA. Tel: +1 518 537 4700, Fax: +1 518 537
5899, Email: psc@periodicals.com
Journal of Business Logistics, (Print ISSN: 0735-3766 Online ISSN: 2158-1592) is published quarterly on behalf , The Council of Supply Chain
Management Professionals by Wiley Subscription Services, Inc., a Wiley Company, 111 River St., Hoboken, NJ 07030-5774.
Postmaster: Send all address changes to Journal of Business Logistics, Journal Customer Services, John Wiley & Sons Inc., 350 Main St., Malden,
MA 02148-5020.
Journal Customer Services: For ordering information, claims and any enquiry concerning your journal subscription please go to www.wiley
customerhelp.com/ask or contact your nearest ofce.
Americas: Email: cs-journals@wiley.com; Tel: +1 781 388 8598 or +1 800 835 6770 (toll free in the USA & Canada).
Europe, Middle East and Africa: Email: cs-journals@wiley.com; Tel: +44 (0) 1865 778315.
Asia Pacific: Email: cs-journals@wiley.com; Tel: +65 6511 8000.
Japan: For Japanese speaking support, Email: cs-japan@wiley.com; Tel: +65 6511 8010 or Tel (toll-free): 005 316 50 480.
Visit our Online Customer Get-Help available in 6 languages at www.wileycustomerhelp.com
Wileys Corporate Citizenship initiative seeks to address the environmental, social, economic, and ethical challenges faced in our business and
which are important to our diverse stakeholder groups. Since launching the initiative, we have focused on sharing our content with those in need,
enhancing community philanthropy, reducing our carbon impact, creating global guidelines and best practices for paper use, establishing a vendor
code of ethics, and engaging our colleagues and other stakeholders in our efforts.
Follow our progress at www.wiley.com/go/citizenship
View this journal online at wileyonlinelibrary.com
Access to this journal is available free online within institutions in the developing world through the AGORA initiative with the FAO, the
HINARI initiative with the WHO and the OARE initiative with UNEP. For information, visit www.aginternetwork.org, www.healthinternetwork.org,
www.oaresciences.org
Printed in the USA by The Sheridan Group
jbl_34_2_prelim.indd 1
5/23/2013 7:49:09 AM
University of Arkansas
Weber State University
e illuminate the myriad of opportunities for research where supply chain management (SCM) intersects with data science, predictive analytics, and big data, collectively referred to as DPB. We show that these terms are not only becoming popular but are also relevant to
supply chain research and education. Data science requires both domain knowledge and a broad set of quantitative skills, but there is a dearth
of literature on the topic and many questions. We call for research on skills that are needed by SCM data scientists and discuss how such skills
and domain knowledge affect the effectiveness of an SCM data scientist. Such knowledge is crucial to develop future supply chain leaders. We
propose denitions of data science and predictive analytics as applied to SCM. We examine possible applications of DPB in practice and provide examples of research questions from these applications, as well as examples of research questions employing DPB that stem from management theories. Finally, we propose specic steps interested researchers can take to respond to our call for research on the intersection of SCM
and DPB.
Keywords: data science; predictive analytics; big data; logistics; supply chain management; design; collaboration; integration; education
INTRODUCTION
Big data is the buzzword of the day. However, more than the
typical faddish fuzz, big data carries with it the opportunity to
change business model design and day-to-day decision making
that accompany emerging data analysis. This growing combination of resources, tools, and applications has deep implications in
the eld of supply chain management (SCM), presenting a doozy
of an opportunity and a challenge to our eld. Indeed, more data
have been recorded in the past two years than in all of previous
human history.1 Big data are being used to transform medical
practice, modernize public policy, and inform business decision
making (Mayer-Schonberger and Cukier 2013). Big data have
the potential to revolutionize supply chain dynamics.
The growth in the quantity and diversity of data has led to
data sets larger than is manageable by the conventional, handson management tools. To manage these new and potentially
invaluable data sets, new methods of data science and new applications in the form of predictive analytics, have been developed.
We will refer to this new conuence of data science, predictive
analytics, and big data as DPB.
Data are widely considered to be a driver of better decision
making and improved protability, and this perception has some
data to back it up. Based on their large-scale study, McAfee and
Brynjolfsson (2012) note, [t]he more companies characterized
themselves as data-driven, the better they performed on objective
measures of nancial and operational results companies in the
Corresponding author:
Matthew A. Waller, Sam M. Walton College of Business, WCOB
308, University of Arkansas, Fayetteville, AR 72701-1201, USA;
E-mail: MWaller@walton.uark.edu
1
top third of their industry in the use of data-driven decision making were on average, 5% more productive and 6% more protable than their competitors (p. 64). To make the most of the
big-data revolution, supply chain researchers and managers need
to understand and embrace DPBs role and implications for supply chain decision making.
DATA SCIENCE, PREDICTIVE ANALYTICS, AND BIG
DATA
There is growing popular, business, and academic attention to
DPB. For instance, the October 2012 issue of Harvard Business
Review contained three articles that are relevant to this editorial:
Big Data: The Management Revolution (McAfee and Brynjolfsson 2012), Data Scientist: The Sexiest Job of the 21st
Century (Davenport and Patil 2012), and Making Advanced
Analytics Work for You (Barton and Court 2012). MIS Quarterly had a special issue on business intelligence and the lead
article was titled, Business Intelligence and Analytics: From Big
Data to Big Impact (Chen et al. 2012). There is also a plethora
of articles in trade and even lay publications on these topics.
There is even a new journal, Big Data, which premiered in
March 2013.
Over the past few years, we have been trying to understand
the DPBs implications for research and education in business
logistics and SCM. We believe that these new tools will transform the way supply chain are designed and managed, presenting
a new and signicant challenge to logistics and SCM. Meeting
this challenge may require changes in foci of research and education. Many traditional approaches will need to be re-imagined.
Some standard practices may even be discarded as obsolete in
the new data-rich environment. Some may see the possibilities as
threats rather than opportunities. Yet DPB and SCM are fundamentally compatible, thus the tremendous value of DPB lies
within our grasp.
78
Narrow Set of
Analy cal Skills
Domain Knowledge
It is tempting to think that this is the difference between marketing and marketing science, however, most PhDs in marketing
science tend to have a very strong focus on one methodological
area.
79
Table 1: Examples of skills needed by a supply chain management (SCM) data scientist
SCM data scientist skill set
Discipline
More important
Less important
Statistics
Forecasting
Optimization
Discrete event
simulation
Applied probability
Analytical mathematical
modeling
Finance
Economics
Marketing
Accounting
80
Dimension
of interest
Statistics
Quantitative
Forecasting
Optimization
Minimization and
maximization
Discrete event
simulation
Quantitative analysis
of a system in a
stochastic setting
Description of
stochastic variables,
expected values,
and uncertainty
Search for patterns
and relationships
between a large
number of variables
with lots of data
Precise analysis using
articial and unrealistic
assumptions for
theorems and proofs
Applied probability
Data mining
Analytical
mathematical
modeling
Relevant
Less relevant
Asymptotic properties of
Gaussian processes
81
Consumer
Inventory
Location and time
Volume
Velocity
Big data
Big data is unique because of the volume, variety, and velocity
of the data, which today is widely available and much less
expensive to access and store (McAfee and Brynjolfsson 2012).
Volume can occur in many ways. There are more data because,
among other reasons, the data are captured in more detail. For
instance, instead of just recording that a unit sold at a particular
location, the time it was sold and the amount of inventory at the
time of the sale, is also captured. As another example, many
companies that did not record daily sales by location and by
stock-keeping unit to make inventory decisions, now do. Moreover, long global supply chains necessitate data capture at multiple points in the supply chain. In addition, there is now a
proliferation of consumer sentiment data resulting from Tweets,
Likes, and product reviews on websites. Such data must be analyzed and quantied. Software companies that provide algorithms
designed to assess text from reviews and Tweets are cropping up
Variety
http://www.google.com/trends/explore#q=%22data%20science
%22%2C%20%22data%20scientist%22&cmpt=q (referenced March
22, 2013).
82
User
Inventory
management
Forecasting
Transportation
management
Human
resources
Carrier
Time of delivery,
factoring in weather,
driver characteristics,
time of day and date
Optimal routing,
taking into account
weather, trafc congestion,
and driver characteristics
Manufacturer
Early response
to extremely negative
or positive customer
sentiment
Reduction in shrink,
efcient consumer
response, quick response,
and vendor managed
inventory
Improved notication of
delivery time, and
availability; surveillance data
for improved yard management
Retailer
Customer sentiment
data and use of
mobile devices in stores
Improvement in
perpetual inventory
system accuracy
Reduction in
driver turnover,
driver assignment,
using sentiment
data analysis
More effective
monitoring of
productivity;
medical sensors
for safety of labor
in factories
Reduction in labor
due to reduction in
misplaced inventory
Type of data
Sales
Consumer
Inventory
management
Transportation
management
increased need for people with skill sets that can deal with big
data. Figure 3B shows a graph of searches5 for Predictive Analytics. Like Data Science, searches for Predictive Analytics
essentially began in 2005 with signicant growth after 2009. Fig5
http://www.google.com/trends/explore#q=%22predictive%
20analytics%22&cmpt=q (referenced March 22, 2013).
http://www.google.com/trends/explore#q=%22big%20data%
22%2C%20%22supply%20chain%20management%22&cmpt=q
(referenced March 22, 2013).
83
Table 6: Examples of big data research questions that are relevant to supply chain management (SCM), stemming from management
theory
Theory
Research question
Transaction
How does the existence of big data affect the reduction in internal transaction costs vis-"a-vis external
cost economics
transaction costs and how is this affecting the size of logistics organizations and the structure of supply chains?
Resource-based view Can SCM data science be developed as a resource that is valuable, rare, in imitable, and nonsubstitutable?
Contingency theory How can big data and SCM data science be used by logistics managers to meet internal needs and adjust to
changes in the supply chain environment?
Resource
How does the ability to use big data for SCM decisions affect a rms power in comparison with its suppliers
dependence theory
or customers?
Agency theory
How does the proliferation of big data affect the agency costs associated with the use of third party logistics?
Institutional theory
How do differences in freedom of information between countries affect rms operating under these different
institutions in terms of their abilities to leverage big data in the supply chain?
Panel C: Google Searches for Big Data and Supply Chain Management since 2004
84
ACKNOWLEDGMENTS
CONCLUDING DISCUSSION
Although Big data has become a contemporary buzzword, it has
signicant implications in our discipline, and presents an opportunity and a challenge to our approach of research and teaching. We
can easily see how data science and predictive analytics apply to
SCM, but sometimes nd it more difcult to see the direct connection of big data to SCM. Therefore, we would like to see research
published in the Journal of Business Logistics that brings clarity to
the relevance of big data, and DPB in general within the supply
chain domain. Here is how you can participate:
1.
2.
3.
4.
5.
orty years have passed since the rst multi-item measurement scales were employed through survey research to better understand important
logistics concepts. Through the years, four leading logistics journals have published research containing a total of 1,670 scales within 283
articles. A 42% increase in utilization has occurred during the most recent decade. The research fully discloses the conceptualization, composition, and properties of the multi-item survey scales utilized in the study of logistics and supply chain management theory and practice. By documenting each scale published from 2001 to 2010, the authors make comparisons with results from the 19732000 study. Primary ndings
indicate an increase in the percentage of research utilizing multi-item scales, and an increase in the application of conrmatory factor analysis
(CFA) has assisted in producing more highly acceptable psychometric properties of the scales. This implies that researchers are utilizing more
highly developed scales and following strict scale-development procedures for building more robust scales to measure concepts important for
advancing our knowledge of logistics and supply chain management. The top four conceptual categories for the scales have remained the same
since 1973 and account for greater than half of the scales published. Eight additions to the categories since 2001 pertain greatly to controlling
our supply chains for customer, brand, and overall security benets. The compendium of scales provides a central document to reference as
researchers seek to employ highly developed survey measures.
Keywords: survey; measurement; scales; logistics; metrics; reliability; supply chain management
Corresponding author:
Scott Keller, Professor, Logistics and Marketing, University of West
Florida, 11000 University Parkway, Pensacola, FL 32514-5750,
USA; E-mail: skeller@uwf.edu
The research followed the method for identifying and categorizing the scales that was employed by Keller et al. (2002) and
will be discussed in the next section. Results and observations
then will be discussed to disclose scale characteristics and the
86
S. B. Keller et al.
87
Table 1 indicates the total number of scales reported in the articles from 2001 to 2010 to be 980 scales. Comparing this to the
rst three decades of logistics research (690 scales reported), the
discipline has achieved a 42% increase within the past 10 years,
alone [(980690)/690]. It is clear that researchers in logistics and
supply chain management have found survey research employing
multiple items increasingly useful for better capturing the conceptual domains of constructs not directly measurable utilizing
ratio metrics.
Scale categorizations
Table 1 lists the categories utilized by the authors to offer
broad categorization of the scales. First, scale descriptions and
items provided in each article were reviewed and categorized
by each author and discrepancies between authors were discussed and resolved. Four iterative rounds were necessary to
come to full agreement on the categorization of the scales. It is
important to note, however, that it was not our intention to
completely content analyze each scale and the conceptualization
provided in the research; rather, the categorization is used to
organize the scales in a manageable manner so that researchers
in the future may peruse categories to locate scales useful to
their research. Second, the authors started grouping scales under
the categories resulting from Keller et al. (2002). Scales were
added, some groupings changed, categories were added, and
some dropped.
Performance and Quality contained 188 scales, accounted for
19.2% of the total scales reported, and ranked rst in our categorization. In comparison, Keller et al. (2002) reported this category as ranked second with 117 scales totaling 17% of the total
scales. Examples of this category include:
Consequence (Inbound) (Svensson 2002); The negative consequences of disturbances, that is, the degree of negative impact
in the rms inbound logistics ows. Sample item: How often
or seldom do the disturbances in your companys inbound
logistics ows from subcontractors cause downtimes in the
manufacturing processes?
Financial Performance (Corsten and Felde 2005); Return on
assets (ROA), return on sales, and the improvement of both
measures compared to the industry average. Sample item:
Compared to the average in our industry our ROA has been
considerably better over the last three years.
Channel Relationships ranked second in this study (184 scales;
18.8% of total scales reported). Again, the percentages were similar to the results in Keller et al. (2002) with 17.4% of the total
reported scales, but the category ranked rst from 1973 to 2000.
Examples of this category include:
Altruism (Interorganizational Citizenship Behaviors) (Autry
et al. 2008); Behavior directed at helping a partner rm in
solving problems or acquiring needed skills/knowledge. Sample item: When our supplier is unable to fulll certain responsibilities, we try to be understanding.
Attachment (Knemeyer and Murphy 2005); Describes genuine
feelings toward the other company or that companys employees. Sample item: In times of need, this third party has gone
out on a limb for us.
Completing rankings three and four, Internal Relationships/
Personnel (third) and Information Sharing, Processes, and Technology (fourth) remained in their respective positions across both
studies. While positions one and two swapped categories, the top
Category
# of scales
% of total
Cumulative %
188
184
133
113
52
44
37
26
24
24
21
20
20
20
20
15
11
10
10
8
980
19.2
18.8
13.6
11.5
5.3
4.5
3.8
2.7
2.4
2.4
2.1
2.0
2.0
2.0
2.0
1.5
1.1
1.0
1.0
0.8
19.2
38.0
51.5
63.1
68.4
72.9
76.6
79.3
81.7
84.2
86.3
88.4
90.4
92.4
94.5
96.0
97.1
98.2
99.2
100.0
88
S. B. Keller et al.
Number of Articles
30
66
25
57
20
Total
IJLM
15
IJPDLM
32
JBL
10
TJ
13
6
5
1
0
1971-75 1976-80 1981-85 1986-90 1991-95 1996-00 2001-05 2006-10
Year
*The bar chart is on a separate scale from the line charts.
four categories were the same for both studies and accounted for
more than half of the scales reported (200110 = 63.1% cumulative; 19732000 = 56.4% cumulative).
Examples of Internal Relationship/Personnel include:
Behavior Management (Change Process-Refreezing Phase)
(Greer and Ford 2009); Consists of activities that provide
feedback on individual performance and incentives for acting
in a manner that promotes and reinforces desired behavior
during change implementation. Sample item: Were employees rewarded for working to support the change effort?
Centralization (Diffusion of Contingency Planning Process)
(Skipper et al. 2009); Degree of decision-making concentration. Sample item: Participation of subordinates in organizational decision making is encouraged.
Examples of Information Sharing, Processes, and Technology
include:
Adopting Information Technology (Zacharia and Mentzer
2004); The utilization of information technology in the industry. Sample item: How extensively is real-time product tracking used in your industry?
Business-to-Business (B2B) E-Commerce Implementation
(Iyer et al. 2004); Encompasses electronic data and information exchange (EDI) technology and application tools such
as the Internet and EDI that facilitate integration and management of core businesses processes between key supply
chain partners. Sample item: Rate the extent to which B2B
e-commerce has been used to generate cross-rm process
integration with key trading partners
Table 1 also indicates the category Customer Satisfaction and
Loyalty jumped from 13th (11 scales and 1.6%) (Keller et al.
2002) to 5th (52 scales and 5.3%) in this study.
Benevolence (Hofer et al. 2009); Extent that the customer
believes that the 3PL has intentions and motives benecial to
89
220
200
Number of Scales
180
417
160
350
140
120
IJLM
100
IJPDLM
192
80
JBL
60
40
20
Total
35
4
TJ
69
40
0
1971-75 1976-80 1981-85 1986-90 1991-95 1996-00 2001-05 2006-10
Year
*The bar chart is on a separate scale from the line charts.
90
S. B. Keller et al.
Figure 3: Range and average number of items per scale in leading logistics journals.*
8
16
Number of Items
19
18
17
16
17
15
Range
IJLM
IJPDLM
JBL
TJ
Combined
3
3
2
.93
.98
.97
.97
.91
.87
Coefficient Alpha
0.85
0.8
Range
Combined
0.75
IJLM
0.7
IJPDLM
JBL
0.65
0.6
TJ
.64
.60
0.55
.54
0.5
1976-80
1981-85
1986-90
.52
1991-95
.53
1996-00
.51
.54
2001-05
2006-10
Year
*Coefficient alpha was first introduced in 1977. Combined includes the average of IJLM, IJPDLM, JBL, and TJ.
The bar chart and line chart are on the same scale.
91
Table 2: Percentage of articles reporting validity and reliability assessments in leading logistics journals (200110)
IJLM
Pretest
Factor/Principal components analysis
Coefcient Alpha/Kuder Richardson 20 reliability
Item-to-total correlations
Conrmatory factor analysis
Average variance extracted compared to shared variance
Total number of articles
(57)
(64)
(50)
(7)
(14)
(0)
(14)
IJPDLM
47
47
75
6
63
22
32
(38)
(64)
(56)
(4)
(9)
(0)
(45)
42
50
73
8
42
2
52
JBL
(58)
(93)
(75)
(28)
(23)
(8)
(40)
68
65
88
14
78
25
57
TJ
(41)
(53)
(65)
(6)
(0)
(0)
(17)
Combined
38
62
77
12
46
0
26
(47)
(72)
(63)
(13)
(13)
(3)
(116)
52
56
79
10
59
13
167
Coefcients within parentheses were reported by Keller et al. (2002) for 19712000.
(27)
(167)
(4.92)
(.81)
57
356
4.17
.83
JM
(63)
(505)
(4.48)
(.81)
159
1,356
3.99
.89
(59) 68
(81) 65
(60) 57
(46) 23
(93) 88
(89) 91
(33) 14
(26) 77
(11) 25
(22) 8
(65) 72
(32) 48
92
cance, magnitude, and direction of relationships among constructs within a theoretical model. It is clear that logistics and
supply chain researchers have amassed a robust foundation of
multi-item measures that have important valid and reliable characteristics. Researchers employing the scales for empirical testing
of hypotheses may have greater condence that their results are,
indeed, a function of the theory and not an anomaly of the measures.
Researchers of performance, quality, relationships (internal
and external), information sharing, processes, and technology
have the greatest bank of scales developed and are encouraged to
continue utilizing and rening the strength of these 618 scales.
Although these topics represent 63% of the cumulative total
scales over the most recent decade, it should also be indicative
of the topics of greatest interest to researchers, editors, reviewers,
and practitioners.
Topics new to the scale categories from that published in Keller et al. (2002) include regulation, safety, customer orientation,
security, supplier selection, reverse logistics, brand image, and
education. These represent a need for further exploration into the
concepts important for building knowledge specic to each area.
Perhaps a common element among the new categories is the
need to better develop our knowledge pertaining to controlling
our supply chains for customer, brand, and overall security benets. Knowledge within these topics is in greater need of exploration and development.
The driving force for this research was to provide researchers
a central document to reference when searching for developed
measures to utilize in future studies. The compendium of scales
appearing in Appendix S1 accomplishes this goal. It also may
assist in identifying additional articles that are related to the concepts and theories of immediate importance to researchers.
Along with studies of journal rankings and impact factors, our
comparisons of the statistics related to the multi-item scales utilized in JBL and JM over the most recent decade (200110) offer
signicant indications of the high quality and rigor of the
research published in JBL. It is hoped that the entire results of
this study further assist professors in justifying the strength of
our top logistics journals to our peers, chairpersons, deans, and
others that require support for rendering tenure and promotion
decisions.
REFERENCES
Autry, C.W., and Grifs, S.E. 2005. A Social Anthropology of
Logistics Research: Exploring Productivity and Collaboration
in an Emerging Science. Transportation Journal 44(4):2743.
Autry, C.W., Skinner, L.R., and Lamb, C.W. 2008.
Interorganizational Citizenship Behaviors: An Empirical
Study. Journal of Business Logistics 29(2):5374.
Cahill, D.L., Goldsby, T.J., Knemeyer, A.M., and Wallenburg, C.M.
2010. Customer Loyalty in Logistics Outsourcing Relationships:
An Examination of the Moderating Effects of Conict
Frequency. Journal of Business Logistics 31(2): 25377.
Carter, C.R. 2002. Assessing Logistics and Transportation
Journals: Alternative Perspectives. Transportation Journal
42(2):3950.
S. B. Keller et al.
93
lthough agility has been identied as one of the most important issues of contemporary supply chain management, the theoretical basis
for understanding supply chain agility is fragmented. This research addresses the gap related to the ambiguity surrounding the dimensions
and denitions of rm supply chain agility by employing a multidisciplinary literature review to gain an in-depth understanding of agility. In
addition, a comprehensive measurement instrument that draws on the foundations of social and life science theory is developed and empirically
validated. The results of the research indicate that rm supply chain agility is composed of ve distinct dimensions including alertness, accessibility, decisiveness, swiftness, and exibility. Based on these elements, a comprehensive denition of rm supply chain agility is developed for
further theoretical testing of the concept.
Keywords: agility; rm supply chain agility; supply chain agility; scale development
INTRODUCTION
Agility has emerged as the dominant competitive vehicle for
organizations operating in uncertain and ever-changing business
environments, and has been heralded as the business paradigm of
the 21st century (Tseng and Lin 2011). The concept has risen in
signicance as businesses no longer compete as solely autonomous entities, but rather as supply chains (Lambert and Cooper
2000; Christopher 2005; Defee and Stank 2005; Stank et al.
2005). Supply chain members must be capable of rapidly aligning their collective capabilities to respond to changes in demand
and supply (Gligor and Holcomb 2012a). Furthermore, it has
been recognized that to achieve a competitive advantage in the
rapidly changing business environment, rms must align with
suppliers and customers to coordinate operations and together
achieve a level of agility beyond that of competitors (Lin et al.
2006). As supply chain agility has progressed from a conference
topic to a practical imperative for most companies (White et al.
2005), agility has been highlighted as the fundamental characteristic of the best supply chains (Lee 2004).
Although agility has been identied as one of the most important issues of contemporary supply chain management (Lee
2004), the theoretical basis for understanding supply chain agility
is fragmented (Li et al. 2008). Agility is a broad and multidimensional concept bridging many disciplines (Gligor and
Holcomb 2012a). The multidimensionality of agility has led to
much confusion and ambiguity (Giachetti et al. 2003; Li et al.
2009; Gligor and Holcomb 2012a). Elements and linkages
among agility elements are underdeveloped, and it is uncommon
for any two articles to adopt the same denition (Conboy 2009).
A rigorously validated instrument to measure supply chain agility
Corresponding author:
David M. Gligor, University of Missouri-Kansas City, Department
of Marketing and Supply Chain Management, Henry W. Bloch
School of Management, 5100 Rockhill Rd., Kansas City, MO
65110, USA; E-mail: mars2u7@yahoo.com
is needed to enable researchers to credibly test explanatory theories regarding causal links among capabilities, practices, and performance outcomes related to this phenomenon (Sherehiy et al.
2007; Li et al. 2009).
This research addresses the gap related to the ambiguity surrounding the dimensions and denitions of rm supply chain
agility by employing a multidisciplinary literature review to gain
an in-depth understanding of agility. Specically, the sports
science and military science theoretical bases are investigated to
better understand agility and identify its dimensions, and dene
it in a rm supply chain context. Furthermore, a comprehensive
measurement instrument that draws on the foundations of social
and life science theory is developed and empirically validated.
LITERATURE REVIEW
Scholars from multiple business disciplines have dened agility
in ways that emphasized different aspects of the concept. Gligor
and Holcomb (2012a) note that to date, many different denitions and characterizations of agility have appeared in the business literature. Through their comprehensive examination of the
literature, the authors found that the denition and concept of
agility is evolving. For example, much of the earlier research
focused on agility as an ability that enabled rms to thrive in an
environment of continuous and often unanticipated change (Gunasekaran 1998, 1999; Dove 1999; Shari and Zhang 1999; Sarkis 2001). More recently, Vinodh (2010) conceptualizes agility
as a paradigm that facilitates companies to quickly respond to
customers dynamic demands. The concept, which initially
concentrated on manufacturing, has expanded to become a wideranging response to a myriad of business challenges in a
turbulent environment (Yauch 2011; Zhang 2011). Despite the
evolution of the concept, inconsistencies in the multiple business
denitions of agility have been further manifested in the existing
supply chain research in its treatment of agility as a rm concept.
As Gligor and Holcomb (2012a) indicate, few researchers
95
players initiate a change in direction movement before the opponents ball release due to anticipation of the other players movements (Sheppard and Young 2006). Visual search and
anticipation research have also shown that highly skilled athletes
are able to successfully predict the action of an opponent before
it is carried out (Bradshaw et al. 2010). The national protocol for
the assessment of agility performance in team-sport athletes also
recognizes the role of alertness and suggests that the athletes
ability to successfully use agility maneuvers in the actual game
depends on factors such as visual processing, timing, reaction
time, perception, and anticipation (Ellis et al. 2000).
Various conceptualizations of alertness have been introduced
in military science. Dekker (1999) sees agility as the ability to
perceive an upcoming threat and respond to it quickly, while the
U.S. Army denes it more simply as the ability of friendly forces
to act faster than the enemy (US Army 1997). It has been suggested that creating an agile military force requires speeding up
the so-called OODA (observe, orient, decide, act) loop (Fewell
and Hazen 2005). The concept of an OODA loop was developed
by military strategist USAF Colonel John Boyd, and was originally applied at the operational and strategic levels in military
combat operations. The alertness dimension of agility is captured
within the observe and orient stages of the loop and is a prerequisite to an agile response. Some military science researchers
refer to the alertness capability as situational awareness, and
describe it as the perception of environmental elements with
respect to time and space (Dekker 2006; Sheffer 2006). The
speed of recognition of environmental elements is considered
critical (Alberts 2007). In combat, military forces require early
awareness of upcoming threats. The quicker changes are
detected, the sooner the response can be deployed.
The dimension of alertness has also been a focus of business
agility research. Shari and Zhang (1999) recognize that agile
organizations need a basic ability that consists of sensing, perceiving, and anticipating changes in the business environment.
Zhang and Shari (2000) divide agility capabilities into four
major categories: responsiveness (ability to identify, respond to,
and recover from changes quickly, reactively or proactively),
competency (ability to efciently and effectively realize enterprise objectives), exibility/adaptability (ability to implement
different processes and apply different facilities to achieve the
same goals), and speed (ability to complete an activity as quickly
as possible). Although it introduces some of the possible dimensions of agility, Zhang and Sharis (2000) conceptualization is
problematic. One limitation of this conceptualization is the lack
of distinction between the ability to detect changes and the ability to respond to changes. These two distinct capabilities are
grouped under the responsiveness umbrella. This research
expands on Zhang and Sharis (2000) work and posits alertness
as a distinct dimension of agility. Other research articles also recognize the role of alertness in the design of agile manufacturing
systems (Goldman et al. 1995; Almahamid et al. 2010; Inman
et al. 2011; Vinodh and Prasanna 2011; Zhang 2011).
The role of alertness in achieving the desired level of agility is
also emphasized within the information systems and information
systems development research. Sarker and Sarker (2009) argue
that agility lies in environmental scanning and sense-making
routines for anticipating and recognizing possible or imminent
crises, whereas other authors emphasize the important role of
96
sensing market opportunities and threats (Tallon and Pinsonneault 2011; Tseng and Lin 2011; Lu and Ramamurthy 2012).
Within a supply chain management context, Christopher (2000)
was the rst to acknowledge that, to be truly agile, a supply
chain must be capable of reading and responding to real
demand. He refers to this capability as market sensitivity. One
limitation of Christophers (2000) interpretation is that although
he recognizes the importance of reading customers requirements, he does not conceptualize it as a distinct capability; he
places it in the same category with the responding to real
demand capability. Another drawback of Christophers (2000)
research is that it only recognizes the importance of reading
demand information, with no reference to supply.
Other supply chain researchers also recognize that agility
requires a timely awareness of change and adopt the market sensitivity dimension introduced by Christopher (Lin et al. 2006;
Agarwal et al. 2007; Jain et al. 2008). However, it was Li et al.
(2008) that rst conceptualized alertness as a distinct dimension
of supply chain agility. These authors argue that agile supply
chains must be alert to changes, within the supply chain itself
and within the surrounding environment. This dimension of agility manifests itself through sensing emerging market trends,
listening to customers, and monitoring real demand through daily
point-of-sale data (Li et al. 2008, 2009).
Accessibility
Accessibility emerged from the literature review as the second
dimension of rm supply chain agility. It is dened in this study
as the ability to access relevant data. Research suggests that
once a change is detected through the alertness capability, rms
must also be able to access relevant data to decide how to provide an agile response (Gunasekaran 1998; Sharp et al. 1999;
Jain et al. 2008; Tseng and Lin 2011; Vinodh and Prasanna
2011; Lu and Ramamurthy 2012).
Supply chain-wide information access is recognized as a key
requirement for supply chain agility (Vinodh and Prasanna 2011;
Gligor and Holcomb 2012b). In his seminal article, Christopher
(2000) argues that agile supply chains must possess a number of
distinguishing characteristics. Agile supply chains must be virtual, that is, they must be information-based rather than inventory-based. Supply chain members must share real-time demand,
inventory, and production information (Ahn et al. 2012). The creation of virtual supply chains allows all supply chain members to
access relevant data and make informed decisions about how to
respond to changes detected in the environment. Lin et al. (2006)
refer to the capacity to access information as information integration, and describe it as the ability to use information technology
to share data between buyers and supplies. Information integration can be considered as the infrastructure needed to create a
virtual supply chain (Christopher et al. 2004; Jain et al. 2008).
Manufacturing research also suggests that a requirement for
designing agility is the creation of an environment where relevant information can be accessed. Goldman et al. (1995) consider the formation of virtual partnerships to be one of the four
primary principles of agility. This perspective is supported by
other manufacturing research articles that identify virtual enterprises, information technology, and communication as the key
enablers of agility (Gunasekaran 1998; Sharp et al. 1999; Khalil
D. M. Gligor et al.
enablers as they help offensive players successfully evade opponents (Bradshaw et al. 2010).
In their denition of agility, Young et al. (2002) recognize
that the two main components of agility are change in direction
speed and decision-making factors. Other agility conceptualizations also acknowledge the contribution of decision-making abilities to agility performance in sports (Chelladurai 1976;
Abernethy et al. 1999; Sheppard and Young 2006). Research
also suggests that, as the complexity of the task increases, decision-making skills become more important (Sheppard and Young
2006). The increase in complexity affects an athletes performance as evidenced by the weak correlation between straight
sprinting ability and the ability to perform complex agility tasks
(Tsitskarsis et al. 2003). The decision-making component of agility can help explain why straight sprinting performance (no decision making required) has little to do with agility performance.
Previous research has observed less than 50% commonality
between reactive (decision required) and preplanned (no decision
required) agility performance (Farrow et al. 2005).
In a supply chain context, Christopher (2000) makes a clear
distinction between speed (meeting customer demand through
shortened delivery lead times) and agility (responding quickly to
changes in demand in terms of both volume and variety). Military science research also recognizes the importance of decisiveness. The decide phase is one of the components of the OODA
loop (Fewell and Hazen 2005). A three-step sequential process
takes place during the decide phase: options generation, best
option selection, and best option adaptation. Speeding up the
decide phase is suggested to result in a more agile response
(Dekker 2006).
The above literature review indicates that to develop supply
chain agility, it is not enough to create the abilities to quickly
detect changes (alertness) and access relevant information on
how to deal with changes (accessibility). Firms must also foster
the ability to make resolute decisions on how to respond to
changes (decisiveness). Combined, the alertness, accessibility,
and decisiveness dimensions of agility form the cognitive area of
rm supply chain agility. These dimensions are related to information-processing and allow the rm to determine what actions
to take in response to changes, opportunities, or threats.
Swiftness
Once a decision is made on how to respond to changes, entities
must be able to quickly implement those decisions (Sharp et al.
1999; Gunasekaran and Yusuf 2002; Lin et al. 2006; Alberts
2007; Jain et al. 2008; Mackley et al. 2008). Swiftness, the
fourth dimension of agility, is dened as the ability to implement
decisions quickly. Sports and military science research recognize
the enabling role of swiftness in fostering agility. Research on
the effects of agility training on athletic power performance indicates that agility is highly dependent on the athletes speed of
movement (Sporis et al. 2010). Various sports agility tests have
also identied change in direction speed as one of the pivotal
components of agility (Young et al. 2002; Farrow et al. 2005).
Although the terminology might vary across research articles
(e.g., quickness, rapidness, swiftness, speed, velocity), a majority
recognize swiftness as an essential component of agility (Clarke
1959; Mathews 1973; Draper and Lancaster 1985; Bloomeld
97
98
range. For example, the rms supply chain cannot quickly produce more items than its xed manufacturing capacity allows.
Supply chain agility literature recognizes the role of exibility
in providing an agile response. Empirical research found a direct
positive relationship between procurement and manufacturing
exibility and supply chain agility (Swafford et al. 2006). In
their framework, Swafford et al. (2006) consider supply chain
agility as an externally focused capability that is derived from
exibility (internally focused competency) in supply chain processes. Research also indicates that supply chain exibility
directly and positively impacts supply chain agility (Swafford
et al. 2008). Other supply chain researchers recognize the role of
exibility. In their denition of supply chain agility, Li et al.
(2008, 2009) consider exibility to be a core aspect of the construct. Similarly, this perspective nds support in a number of
supply chain agility frameworks (Christopher 2000; Lin et al.
2006; Jain et al. 2008).
Flexibility has long been identied as a key agility dimension
across manufacturing research. Agility as a business concept was
rst coined in relation to exible manufacturing systems (Nagel
and Dove 1991). The idea of manufacturing exibility was subsequently extended into a wider business context, and the concept of agility as an organizational trait was born (Christopher
and Towill 2002). The role of exibility in providing an agile
response is highlighted within several agility denitions. Hong
et al. (1996) dene agility as exibility and rapid response to
market demands, whereas Eshlaghy et al. (2010); describe it as
a model that provides exibility. In one of the most referenced
frameworks of manufacturing agility, Shari and Zhang (1999)
propose exibility to be one of the capabilities that an agile organization must possess. This perspective is supported by a number
of empirical research articles within the manufacturing realm
(Yusuf et al. 1999; Gunasekaran and Yusuf 2002; Kumar and
Deshmukh 2006; Eshlaghy et al. 2010; Jacobs et al. 2011;
Costantino et al. 2012).
The rm supply chain agility construct
The examination of previous research also guided the classication of the agility dimensions into two higher echelon categories:
physical and cognitive. Research suggests that swiftness and exibility represent the physical dimensions of rm supply chain
agility; alertness, accessibility, and decisiveness exemplify the
cognitive dimensions of the concept. The cognitive dimensions
of rm supply chain agility are related to information-processing
and help rms to determine what actions to take, while the physical dimensions are related to action-taking and enable rms to
implement those actions (see Table 1).
To clearly establish the relationship between supply chain agility and its dimensions, it is important to determine whether the
supply chain agility construct is reective or formative. Three
theoretical considerations can help distinguish formative models
from reective ones (Coltman et al. 2008). The rst theoretical
criterion is the nature of the construct. In reective models, the
latent construct exists independent of the measures used, whereas
in formative models, the latent construct is determined as a combination of its indicators (Rossiter 2002; Borsboom et al. 2003).
The second theoretical consideration pertains to the direction of
causality between items and the latent construct. In reective
D. M. Gligor et al.
Denition
Type
Cognitive
dimensions
Physical
dimensions
models, variation in the construct causes variation in item measures, whereas in formative models, variation in item measures
causes variation in the construct (Bollen and Kwok-Fai 2000;
Edwards and Bagozzi 2000; Diamantopoulos and Siguaw 2006).
The third theoretical criterion considers the characteristics of the
items used to measure the construct. In reective models, items
are manifested by the construct and share a common theme. In
formative models, items dene the construct and need not share a
common theme (Rossiter 2002; Jarvis et al. 2003). Based on
these theoretical considerations and consistent with prior research
(i.e., Li et al. 2009), rm supply chain agility is operationalized
as a second-order reective construct with the rst-order factors
of alertness, accessibility, decisiveness, swiftness, and exibility
(Figure 1).
METHOD
Following the identication of the dimensions of supply chain
agility, the next phase of the research was to develop and test
scales for each of the factors. The procedures used to develop
and assess the validity of the agility scale are described below.
Scale development and survey design
Scale development followed procedures and guidelines recommended by Churchill (1979), DeVellis (1991), Hinkin (1995),
Mentzer and Flint (1997), and Garver and Mentzer (1999). Each
dimension of the second-order construct is measured by multi-item
scales to increase reliability, decrease measurement error, ensure
greater variability among the survey participants, and improve
Alertness
Accessibility
Decisiveness
Swi ness
Flexibility
99
The unit of analysis for the research is the rm, and the preferred
target respondents were senior-level managers with knowledge of
supply chain processes and activities, and direct involvement in
operational and strategic decision making. Data were gathered
using a nonexperimental survey methodology (Kerlinger and Lee
2000). Specically, the research employed an Internet survey to
collect the necessary data for model testing. The web-based survey approach is appropriate because the population of interest is
business, and coverage issues are not present due to high rates of
computer use and the large sample size (Dillman 2007).
Purposive sampling was employed in the hopes of achieving a
moderate level of external validity and to contribute to the generalizability of results (Cook and Campbell 1979). Potential
respondents were identied from two sources. The rst source of
potential participants was a database of supply chain managers
that comprised the mailing list of the supply chain management
program of a large public university. The database contained
contact information for more than 3,000 managers (name, phone
number, email, and title) from U.S.-based companies in a diverse
set of industries. An email was sent to all contacts in the database requesting participation in the study. The Qualtrics software
indicated that the email was received and opened by 285 respondents, conrming that correct/updated contact information existed
for these managers. Therefore, this sample of 285 respondents
was considered for nal survey testing. To increase response
rate, participants were offered an executive summary of the
research ndings and entered into a rafe for the chance to win
$100.
The second source of potential participants was selected from
the panel members of SurveyMonkey, a large third party marketing rm that specializes in survey data collection. SurveyMonkey
provided contact information for 1,135 senior-level managers of
diverse backgrounds, with the knowledge of supply chain processes and activities, and direct involvement in operational and
strategic decision making who were prequalied to participate in
the study. Although participants were not provided any direct
nancial incentives, SurveyMonkey pledged to donate $.50 to
the charity of the respondents choice, and enter the respondents
into a rafe for the chance to win $100.
Potential respondents from both databases (university supply
chain program and SurveyMonkey) were prequalied using the
procedures suggested by Dillman (2007) and Kerlinger and Lee
(2000). Following the purication of the measurement instrument, the main survey test was sent to the sample of 285 potential respondents selected from the database of the universitys
supply chain program, and the sample of 1,135 prequalied
100
D. M. Gligor et al.
Percentage
<1 year
13 years
35 years
510 years
1015 years
1520 years
20+ years
4
3
10
9
10
12
52
Total
100
Total company
annual sales
Percentage
Type of industry
Percentage
<$250 million
$250 million$500 million
$500 million$1 billion
$1$2 billion
$2$3 billion
$3$5 billion
$5$9 billion
>$9 billion
18
12
10
17
13
10
8
12
Energy/Chemical/Mining
Communications/Media/Entertainment
Retail
Manufacturing-General
Manufacturing-Consumer products
Manufacturing-Aerospace/defense
Manufacturing-High technology
Energy/Chemical/Mining
Financial services/Insurance
Life sciences-Pharmaceuticals
Life sciences-Medical devices
Health managed care
Transportation service provider
Other
2
3
18
19
15
4
4
1
2
3
3
2
8
16
100
100
101
Scale/Item
Alertness
A1
A2
A3
A4
A5
A6
Accessibility
B1
B2
B3
B4
B5
B6
B7
Decisiveness
C1
C2
C3
C4
C5
C6
Flexibility
E1
E2
E3
E4
E5
E6
Swiftness
D1
D2
D3
D4
D5
Cronbach Alpha
for scale
Alpha if item
deleted
Item-to-total
correlation
Mean
SD
Item
loadings
Average variance
extracted
.899
.881
.883
.878
.880
.879
.885
.898
.725
.714
.745
.729
.739
.701
3.23
2.97
3.13
3.04
3.13
3.26
1.236
1.169
1.262
1.234
1.276
1.237
.761
.781
.791
.799
.773
.726
.509
.868
.851
.843
.847
.843
.853
.847
.860
.868
.625
.695
.658
.691
.619
.660
.561
2.95
2.91
3.15
2.90
3.19
2.89
2.83
1.168
1.142
1.346
1.176
1.358
1.176
1.272
.751
.785
.657
.780
.644
.668
.577
.487
.868
.851
.850
.846
.863
.870
.865
.880
.725
.729
.753
.654
.612
.639
2.87
2.96
2.94
2.85
2.76
3.06
1.168
1.208
1.191
1.219
1.223
1.185
.786
.785
.805
.674
.665
.732
.552
.841
.808
.802
.818
.832
.810
.820
.843
.654
.689
.606
.635
.647
.591
2.93
2.85
2.71
3.21
2.59
2.78
1.273
1.210
1.291
1.310
1.164
1.236
.687
.772
.686
.637
.681
.665
.475
.882
.861
.850
.861
.852
.860
.882
.700
.570
.522
.545
.509
3.08
3.13
3.23
3.03
2.95
1.276
1.310
1.274
1.331
1.244
.780
.788
.716
.790
.791
.598
CR
102
D. M. Gligor et al.
Alertness
Accessibility
Decisiveness
Swiftness
Flexibility
Alertness
Accessibility
1
.920
.941
.985
.868
1
.922
.914
.920
Decisiveness
Swiftness
Flexibility
1
.911
1
.994
.923
103
128.489
68.069
41.861
150.759
103.763
120.099
67.401
25.03
80.68
129.996
Managers can use the comprehensive list of dimensions examined in this research to determine what aspects of their operations
and tactics should be improved to enhance the rms supply
chain agility. By evaluating their organizations approaches to the
ve dimensions of supply chain agility, managers can identify
aspects of supply chain management that need to be addressed to
increase the rms supply chain agility. For instance, it could be
the case than an organization excels at quickly identifying
changes in its environment (i.e., alertness), but has suboptimal
decision-making processes, which prevents it from making resolute decisions (i.e., decisiveness). Once managers identify weaknesses associated with either one of the ve dimensions,
corrective actions can be taken to reduce or eliminate these vulnerabilities, and increase the rms level of supply chain agility.
104
D. M. Gligor et al.
APPENDIX B
DELETED SCALE ITEMS
Firm Supply Chain Agility: Alertness
A1 We can notice changes in our environment quicker than our
main competitors. (Newly Developed)
A5 As compared with its main competitors, my company is
faster to discover opportunities in its environment. (Newly
Developed)
A6 My company spots threats in its environment more rapidly
than its main competitors. (Newly Developed)
Firm Supply Chain Agility: Accessibility
B1 My rm can acquire the information it needs to respond to
threats in its environment. (Newly Developed)
B2 We can obtain the information we need to address
opportunities in our environment. (Newly Developed)
B4 My company can access the information it needs to deal
with changes in its environment. (Newly Developed)
B6 Our suppliers communicate relevant information to us.
(Newly Developed)
B7 Our customers share pertinent information with us. (Newly
Developed)
Continued.
REFERENCES
Abernethy, B. 1991. Visual Search Strategies and DecisionMaking in Sport. International Journal of Sport Psychology
22(34):189210.
Abernethy, B., and Russell, D.G. 1987. Expert-Novice Differences
in an Applied Selective Attention Task. Journal of Sport
Psychology 9(4):32645.
Abernethy, B., Wood, J.M., and Parks, S. 1999. Can the
Anticipatory Skills of Experts be Learned by Novices?
Research Quarterly for Exercise and Sport 70(3):31318.
Agarwal, A., Shankar, R., and Tiwari, M.K. 2007. Modeling
Agility of Supply Chain. Industrial Marketing Management
36(4):44357.
Ahn, H.J., Chiderhouse, P., Vossen, G., and Lee, H. 2012.
Rethinking XML-Enabled Agile Supply Chains. International Journal of Information Management 32(1):1723.
Alberts, D.S. 2007. The Future of C2: Agility, Focus and
Convergence. Washington, DC: 12th IC2RTS.
Almahamid, S., Awwad, A., and McAdams, A.C. 2010.
Effects of Organizational Agility and Knowledge Sharing
on Competitive Advantage: An Empirical Study in
Jordan. International Journal of Management 27(3):
387404.
Anderson, J., and Gerbing, D. 1982. Some Methods for
Respecifying Measurement Models to Obtain Unidimensional
Construct Measurement. Journal of Marketing Research 19
(4):45360.
Armstrong, J.S., and Overton, T.S. 1977. Estimating
Nonresponse Bias in Mail Surveys. Journal of Marketing
Research 14(3):396402.
Atkinson, S.R., and Moffat, J. 2005. The Agile Organization:
From Informal Networks to Complex Effects and Agility.
Washington, DC: Information Age Transformation Series.
105
106
D. M. Gligor et al.
107
108
D. M. Gligor et al.
SHORT BIOGRAPHIES
David M. Gligor (PhD University of Tennessee) is an Assistant Professor in the Department of Marketing and Supply Chain
Management at the University of Missouri-Kansas City. David
has published in journals such as the Journal of Business Logistics, Journal of Supply Chain Management, the International
Journal of Logistics Management, Supply Chain Management:
An International Journal, Maritime Economics and Logistics,
and Journal of Transportation Management. Prior to academia,
David spent several years in industry working for major corporations such as General Electric, Ryder Integrated Logistics, and
Hapag-Lloyd.
Mary C. Holcomb (PhD University of Tennessee) is an Associate Professor of Supply Chain Management at the University
of Tennessee. Prior to academia, her professional career involved
some 18 years at the Oak Ridge National Laboratory in transportation research and policy issues for the U.S. Departments of
Energy, Transportation, and Defense. Dr. Holcombs professional
background also includes industry experience with the former
Burlington Northern Railroad, General Motors, and Milliken &
Company. She is a principal researcher in one of the longest running annual studiesLogistics and Transportation Trends and
Issuesthat has been conducted for 21 years. Dr. Holcombs
research has appeared in the Journal of Business Logistics,
Transportation Journal, the International Journal of Logistics
Management, Supply Chain Management: An International Journal, International Journal of Physical Distribution and Logistics
Management, and Supply Chain Management Review.
Theodore P. Stank (PhD University of Georgia) is the Harry
and Vivienne Bruce Chair of Business Excellence and Professor
of Logistics and SCM in the College of Business Administration
at the University of Tennessee at Knoxville. His research focuses
on the strategic implications and performance benets associated
with logistics and supply chain management best practices. He is
author of over 90 articles in academic and professional journals
including Journal of Business Logistics, Journal of Operations
Management, Management Science, Supply Chain Management
Review, and Journal of the Academy of Marketing Science. He is
also co-author of the book 21st Century Logistics: Making Supply Chain Integration a Reality and co-editor of Handbook of
Global Supply Chain Management.
Thomas J. Goldsby1, A. Michael Knemeyer1, Jason W. Miller1, and Carl Marcus Wallenburg2
1
2
moderator is any variable that affects the strength of a relationship between a predictor and an outcome variable. While simple in concept,
the application of moderation analysis can yield profound implications to research conducted in logistics and supply chain management.
Moderation analysis illuminates boundary conditions to purported relationships, providing a deeper perspective on what may, to date, represent
generalizable ndings and commonly held beliefs in the eld. Such ndings prove interesting and enrich our theories. Further, moderation relies
on precise measurement of theoretical constructs in order to avoid attenuation of statistical tests and detect interaction effects. This thought leadership piece seeks to: (1) assert the value of moderation analysis and encourage a more prominent place in our survey-based research projects,
(2) provide best practice approaches for using this type of analysis in pursuit of greater depth and clarity in our research, and (3) provide seeds
for potential research projects that could benet from the use of this type of analysis. Guidance is also provided for reviewers who assess manuscripts featuring moderation.
Keywords: moderation analysis; interactions; empirical methods; measurement; boundary conditions
INTRODUCTION
As the logistics and supply chain management (SCM) disciplines
mature, increased emphasis will be placed on establishing the
boundary conditions of our theories through moderation analysis
(Fawcett and Waller 2011). Identifying limits and boundary conditions improves the precision of our theorizing (Edwards and
Berry 2010), increases the validity of our ndings (Leavitt et al.
2010), and helps ensure that we utilize nontautological theories
given that all falsiable theories must have their limits (Popper
1959; Gray and Cooper 2010). Hall and Rosenthal (1991) assert
that if we want to know how well we are doing in the biological, psychological, and social sciences, an index that will serve
us well is how far we have advanced in our understanding of the
moderator variables of our eld (p. 447). Simply stated, the
exclusion of important moderator variables can lead to overgeneralizations that fail to illuminate the boundary conditions under
which purported relationships exist.
In order to leverage moderation analysis effectively, it is critical to understand the connection between moderation and measurement of our theoretical constructs. Moderation and
measurement are tightly intertwined in that (a) measurement error
can result in extreme attenuation of statistical tests for moderation when regression approaches are used (Aiken and West
1991) and (b) common method variance (CMV) severely deates
interaction effects (Siemsen et al. 2010). In addition to illustrating how intimately moderation analysis is linked with measurement, the objectives of this article are: (1) to assert the value of
moderation analysis and encourage a more prominent place in
our disciplines survey-based research projects, (2) to provide
best practice approaches for using this type of analysis in pursuit
Corresponding author:
Thomas J. Goldsby, Department of Marketing and Logistics, Fisher
College of Business, The Ohio State University, Fisher Hall, 2100 Neil
Avenue,Columbus,OH43210,USA;E-mail:goldsby_2@sher.osu.edu
110
T. J. Goldsby et al.
Control Variables
Outcome (Y)
Xw
j1
bj Kj e:
In Equation (1), assuming unstandardized coefcients, i represents the regression intercept; b1, called a rst-order effect, represents the change in Y due to a one-unit increase in X when M
equals zero; b2 represents the change in Y due to a one-unit
increase in M when X equals zero; and b3 represents the change
in the slope of X on Y due to a one-unit increase of M or, equivalently, the change in the slope of M on Y due to a one-unit
increase of X (Hayes and Matthes 2009; Hayes et al. 2012).2
Moderation is present when b3 is signicantly different from
zero.
The effect of X on Y, termed simple slope by Aiken and West
and denoted as hX, can be calculated as the rst partial derivative
of Y with respect to X. This is shown in Equation (2)3:
hX
@Y
b1 b3 M:
@X
111
112
T. J. Goldsby et al.
Analytical techniques
OLS regression
Logistics regression
Hierarchical linear model
Nonlinear structural equation modeling
Type of moderator
Theorizing
moderation hypotheses
Testing
moderation hypotheses
Interpretation and
presentation of results
(assuming the interaction
is signicant)
113
Panel A: qXM = 0
Reliability of X
0.6
0.7
0.8
0.9
0.6
0.7
0.8
0.9
0.36
0.42
0.48
0.54
0.42
0.49
0.56
0.63
0.48
0.56
0.64
0.72
0.54
0.63
0.72
0.81
0.40
0.45
0.51
0.57
0.45
0.52
0.59
0.65
0.51
0.59
0.66
0.74
0.57
0.65
0.74
0.82
0.49
0.54
0.58
0.63
0.54
0.59
0.65
0.70
0.58
0.65
0.71
0.78
0.63
0.70
0.78
0.85
simple slope of the focal predictor differs from zero are calculated,
and (2) the range of the condence bands for the simple slope indicates the precision of its point estimate (Bauer and Curran 2005;
Miller et al. 2013).
In conducting multigroup analysis, plotting the interaction
effect is not necessary given the interpretability of the structural
coefcients across groups. However, it is important that unstandardized effects be reported for multigroup analysis. For example, imagine that in groups one and two the standardized effect
of X?Y is 0.25, but in group one the standard deviation of X is
2 units and the standard deviation of Y is 4 units, while in group
two the values are 2 and 2, respectively. In such a scenario, the
unstandardized effect in group one is 0.50, and the unstandardized effect in group two is only 0.25. Kim and Mueller (1976)
and Newsom et al. (2003) provide a discussion of this issue in
greater detail, but this simple numerical example demonstrates
that reporting standardized coefcients when the variances of X
and Y are different across levels of M can be problematic.
Given the complexity of moderation analysis and the potential
unfamiliarity of reviewers with how to evaluate such analyses,
we provide a set of guidelines in Table 3 helpful for reviewers
who assess manuscripts featuring moderation. Here, too, we
assume that a single moderator is present in each hypothesized
relationship.
Additional benets of moderation analysis
In the measurement domain, identifying signicant interaction
effects provides evidence that a researcher likely suffers from
minimal contamination due to CMV given that CMV can only
attenuate interaction effects (Evans 1985; Siemsen et al. 2010).
Simulation research by Evans (1985) and subsequent analytical
analysis by Siemsen et al. (2010) nd that CMV stemming from
a single unmeasured method factor cannot inate the parameter
estimate of an interaction. Rather, CMV can heavily deate these
estimates, and Siemsen et al. (2010) note that under realistic conditions deation of 255% could be expected. Two implications
stem from these ndings. First, researchers must be cognizant of
the deation of interaction terms from CMV and design their
surveys following guidelines proposed by Podsakoff et al. (2003)
and Rindeisch et al. (2008) in order to minimize these concerns
for deation. Second, given that simulation studies by Richardson et al. (2009) and Chin et al. (2012) recommend that common
approaches for testing for CMV, particularly the use of an
unmeasured, uncorrelated latent method factor, be discontinued,
moderation analysis may represent one of the best, albeit indirect, approaches to evaluate concerns about CMV biasing
ndings.
PUTTING MODERATION TO WORK
To provide direction for future research utilizing moderation
analysis, we provide several examples of commonly researched
relationships in logistics and SCM in Table 4. Here, we introduce
a theoretically motivated moderating variable for the relationship
and an explanation of the prospective benet for testing such an
effect. By these means, the researcher can explore the boundary
conditions of established knowledge.
114
T. J. Goldsby et al.
Testing
Questions to ask
Have the authors developed a logically consistent series of arguments predicting the sign of the
interaction effect (i.e., positive or negative)?
Have hypotheses predicting the effect of X on Y been developed as conditional effects given
the authors have theorized an interaction?
Have the authors reported whether X, M, and XM are in their raw unit score, mean-centered unit score,
or standardized scores?
Have the authors included all constituent (lower-order) terms necessary for testing the interaction?
If conducting multigroup SEM analysis, have the authors established congural and factorial invariance
prior to testing the invariance of the structural pathway?
Have the authors reported unstandardized effects, as standardized effects should not be reported
since these lack interpretational meaning?
Have the authors interpreted the results properly, such as avoiding statements of main effects?
Have the authors provided a visual representation of the interaction, especially when using product
terms, to aid the audience in interpreting the interaction?
Outcome variable
Customer satisfaction
Customer loyalty
Customer
satisfaction (curvilinear)
Trust
Commitment
Behavioral uncertainty
Information technology
usage for exploitation
Collaboration
Environmental
dynamism
Customer satisfaction
Demand uncertainty
Supplier integration
Performance
Supplier integration
(curvilinear)
Relationship outcomes
Transaction-specic
investments
FINAL THOUGHTS
As the logistics and SCM disciplines mature, moderation analysis will become an increasingly important analytical approach to
establish the boundary conditions of our theories. Developing
and testing moderated relationships will increase the precision
of our theorizing (Edwards and Berry 2010) and allow us to
make more informed recommendations to our practitioner audience. Furthermore, establishing boundary conditions to our theories will allow us to begin to prune the increasingly dense
theoretical landscape in the best tradition of Platts (1964) call
for strong inference by focusing our research efforts on theories
that prove robust to falsication (Leavitt et al. 2010). However,
conducting such analyses requires researchers to develop highly
reliable measurement instruments and data collection protocols
that will minimize the concern of CMV given that both measurement-related issues can severely attenuate these effects. It is
our hope that this thought piece will help clarify any misconceptions concerning moderation analysis and provide researchers
in our disciplines with guidance for testing these effects in the
hopes of nding interesting results (Davis 1971).
REFERENCES
Aguinis, H., and Gottfredson, R.K. 2010. Best-Practice
Recommendations for Estimating Interaction Effects Using
Moderated Multiple Regression. Journal of Organizational
Behavior 31(6):77686.
Aiken, L.S., and West, S.G. 1991. Multiple Regression: Testing
and Interpreting Interactions. Thousand Oaks, CA: Sage
Publications, Inc.
Bagozzi, R.P. 2011. Measurement and Meaning in Information
Systems and Organizational Research: Methodological and
Philosophical Foundations. MIS Quarterly 35(2):26192.
Bauer, D.J., and Curran, P.J. 2005. Probing Interactions in Fixed
and Multilevel Regression: Inferential and Graphical
Techniques. Multivariate Behavioral Research 40(3):373400.
Busemeyer, J.R., and Jones, L.E. 1983. Analysis of Multiplicative
Combination Rules When the Casual Variables Are Measured
With Error. Psychological Bulletin 93(3):54962.
Byrne, B.M. 2009. Structural Equation Modeling With AMOS:
Basic Concepts, Applications, and Programming. 2nd ed.
New York: Routledge Academic.
Cahill, D.L., Goldsby, T.J., Knemeyer, A.M., and Wallenburg, C.M.
2010. Customer Loyalty in Logistics Outsourcing Relationships:
An Examination of the Moderating Effects of Conict
Frequency.Journal of Business Logistics 31(2):25377.
Chen, F.F. 2007. Sensitivity of Goodness of Fit Indexes to
Lack of Measurement Invariance. Structural Equation
Modeling 14(3):464504.
Chin, W.W., Thatcher, J.B., and Wright, R.T. 2012. Assessing
Common Method Bias: Problems With the ULMC
Technique. MIS Quarterly 36(3):100319.
Cohen, J. 1988. Statistical Power Analysis for the Behavioral
Sciences. 2nd ed. Mahwah, NJ: Erlbaum.
Cohen, J., Cohen, P., West, S.G., and Aiken, L.S. 2003. Applied
Multiple Regression/Correlation Analysis for the Behavioral
Sciences. 3rd ed. New York: Routledge.
115
116
T. J. Goldsby et al.
lthough logistics and supply chain management research efforts have continued to become more sophisticated, opportunities still exist for
further advancement. One critical area needing improvement is enhanced scale development and validation. This study provides insights
into the strengths and weaknesses of one specic construct, trust, by examining how trust has been measured in our literature. A methodology
for improving scale development, the Q-sort procedure, is used to demonstrate a need for better measurement scales. Our ndings suggest that
numerous measures of trust have been used for logistics and supply chain researchmost of which have been borrowed from other disciplines
without signicant replication. More importantly, the results illustrate that all too often content validity is less than adequate. It appears as
though we are not always measuring what we want to. Greater focus is recommended for future research that examines theory-based scale
development, purication of measures, and replication of scales to further rene and improve construct measurement for enhanced survey
research.
Keywords: Q-sort; scale development; scale validity; trust
INTRODUCTION
Logistics/supply chain management (SCM) research has evolved
signicantly since the rst issue of the Journal of Business
Logistics appeared in 1978. As an example of this evolution,
Georgi et al. (2010) summarize how the intellectual foundation
of the journal has changed since its inception, shifting from
predominantly operationally focused research to more strategicoriented research. Perhaps much of this evolution is a result of,
or was at least partially motivated by, calls from within the
eld for more rigorous theoretical and methodological focus.
For example, Stock (1990, 1996, 2002) calls for the logistics
discipline to broaden its perspectives on logistics and its role
within the corporate landscape by not only delving into research
in nontraditional areas but also including other disciplinary
approaches, such as psychology and organizational behavior, in
our research. In 1994, Dunn et al. discuss the need for greater
scale development and validation in logistics research as the
eld was embarking on research that involved more latent variables. Scale development and validation was critical to move
logistics research from purely descriptive to a more theoretically
driven approach (Dunn et al. 1994). Furthermore, as research in
logistics began focusing on more behavioral approaches, scale
development and validation were considered more critical for
advancing logistics theory (Keller et al. 2002).
Despite such evolution, recent research has indicated that more
work is needed. Sanders and Wagner (2011) propose that multidisciplinary and/or multimethod research can provide greater
insight into todays and tomorrows challenges and opportunities (p. 317). In addition, Goldsby and Autry (2011, 324) suggest despite the acceleration of SCM research, many critical
topics remain understudied, and indicate that research is war-
Corresponding author:
Judith M. Whipple, Department of Supply Chain Management, The
Eli Broad College of Business, Michigan State University, 632
Bogue Street, East Lansing, MI 48824, USA; E-mail: whipple@bus.
msu.edu
118
be conrmed whenever existing measures are used in substantially different research settings.
Finally, McEvily and Tortoriello (2011) analyze 171 papers
published in the past 48 years in the organizational literature,
and nd 129 different measures of trustfew of which were
strongly replicated and/or supported through construct validity.
As advocated by Goldsby and Autry (2011), SCM researchers
need to focus on greater validation of concepts. Trust represents
a key area in need of such validation. In fact, a new journal was
recently created, the Journal of Trust Research, with a mission
to open the black box of trust (Li 2011, 2). Logistics and SCM
researchers need to be involved in this ongoing debate and study
of trust given that it has been generally accepted that trust is
imperative to constructive interpersonal, interrm and international interaction and cooperation (Li 2011, 1). We should not
continue to rely on trust denitions and scales from the past and
other disciplines without understanding current developments and
enhancements. That is not to say that these denitions and scales
are wrong, per se; however, as a discipline, we need to ensure
that the scales we use are valid and reliable in our context given
the current knowledge base.
The purpose of this study was to review the various denitions and applications of trust from a multidisciplinary perspective, and to examine, specically within the logistics/SCM
literature, how trust has been operationalized from a measurement perspective. We focused on two specic research questions:
(1) what is the state of scale development with respect to trust to
understand how trust has been measured and used in our discipline? and (2) is there consensus with respect to how trust has
been measured and conceptualized to understand issues of scale
validity? As a result of this review and empirical analysis of trust
measures, we propose a framework for examining trust as well
as a future research agenda that can lead to a more thorough,
valid, and enhanced understanding of trust within the logistics/
SCM eld.
LITERATURE REVIEW
Trust has been examined from various disciplinary perspectives,
and numerous authors have suggested multidisciplinary frameworks to tie the varied perspectives together and/or to organize
and review historic perspectives. Our focus here is not to reinvent the wheel, but rather to highlight the various philosophical
traditions associated with trust as well as the current debates
within and across disciplines concerning trust. As discussed by
Rousseau et al. (1998, 393), disciplines have considered trust
from different perspectives:
J. M. Whipple et al.
McEvily (2011) argues that trust has evolved from two dominant paradigms: economics, which emphasizes behavioral measures; and social sciences, which emphasizes attitudinal
measures. Behavioral measures are rooted in economic game
theory and enable risk-based approaches for studying trust,
while attitudinal measures consider trust as a psychological state
(McEvily 2011, 1268). Following this approach, Kramer (1999)
differentiates trust, examined in organizational theory, as being
either a choice behavior (p. 572) or a psychological state (p.
571). As Kramer (1999) concludes, there are two perspectives on
choice behavior: rational choice (e.g., Williamsons 1993 calculative perspective) and relational models of trust (e.g., Granovetters 1985 social embeddedness approach). With respect to
psychological states, various considerations have been examined,
including vulnerability, expectancy, and affective conditions
(Kramer 1999). Li (2007) indicates that, while a majority of trust
researchers adopt a trust as attitude approach, there is growing
interest in the trust as choice approach.
Hosmer (1995, 391) developes the following four categories
of trust based on various research perspectives used, comparing
denitions, illustrating key assumptions, and considering contextual differences: (1) individual expectations (nonrational behavior based upon past experiences and future forecasts); (2)
interpersonal relationships (nonrational behavior, based upon
characteristics and traits of both individuals); (3) economic
exchanges (economically rational behavior, constrained by contracts and controls); and (4) social structures (socially rational
behavior, directed by formal requirements and informal obligations). Hosmer (1995) offers a fth element, ethical principles,
as an important trust consideration.
Hosmers (1995) categorization alludes to trust occurring at
various levels, such as between individuals as well as between
organizations. The level of analysis for studying trust has often
been related to the disciplinary perspective, such that economists
have tended to examine trust at the individual and/or the rm
level, psychologists have focused on the individual and the group
level, and sociologists have examined trust within the group and/
or society (Rousseau et al. 1998). Ganesan and Hess (1997) propose that trust could exist at four levels: interpersonal (i.e., trust
between individuals); organizational (i.e., an individuals trust for
an organization); intraorganizational (i.e., trust between an
employee and employer); and interorganizational (i.e., trust
between organizations).
Specic denitions of trust have developed, in part, from the
perspectives of trust described above. However, researchers still
contend that a widely accepted denition of trust is lacking
(Li 2012, 101). Early denitions of trust considered trust as
either an expectation or a behavioral intention (Moorman et al.
1992). Zucker (1986) offers a denition of trust based on
expectations as indicated: trust is dened as a set of expectations shared by all those involved in the exchange (p. 54).
Expectation approaches indicate that despite uncertainty about
how the other will act, there are positive expectations regarding
the other partys intentions, motivations, and behaviors (Lewicki et al. 2006, 996). Trust as behavioral intention implies a
condence in the other partys motives and, thus, involves risktaking behavior. This behavioral intention entails a willingness
to accept vulnerability to the actions of another party (Lewicki
et al. 2006, 996).
Conceptualizations of Trust
Ganesan (1994) measures two distinct components of interorganizational trust: (1) credibility (one partys expectations
about the required expertise of the other party), and (2)
benevolence (one partys beliefs that intentions and motivations of the other party will be benecial in the face of new
conditions/changes, indicating vulnerability).
Morgan and Hunt (1994) examine trust from an expectation
perspective, considering willingness [to be vulnerable] to be
redundant.
McAllister (1995) measures interpersonal trust from two perspectives: (1) cognitive; and (2) affective. Cognitive trust
involves expectations surrounding reliability and dependability, whereas affective trust entails condent attributions concerning the motivations for that persons citizenship
behavior (McAllister 1995, 30). As such, affective trust may
be considered as a willingness to trust.
Ganesan and Hess (1997) examine credibility and benevolence at two different levels (interpersonal trust between a
buyer and a sales representative and organizational trust that
the buyer and seller have for each others organization)
creating a four-factor model: interpersonal credibility; interpersonal benevolence; organizational credibility; and organizational benevolence.
119
Whipple and Frankel (2000, 23) measure two elements of interorganizational trust, based on Gabarros (1987) intraorganizational work: (1) character-based trust (i.e., integrity,
identication of motives, consistency of behavior, openness,
discreteness); and (2) competence-based trust (i.e., specic
competence, interpersonal competence, competence in business sense, and judgment) trust. Again, trusting in ones character represents elements of vulnerability, whereas trusting in
ones competence represents elements of expectations.
Ganesan and Hess (1997) cite Larson (1992) in their denition of interorganizational trust.
120
J. M. Whipple et al.
Type/level
Character-based
trust
Competence-based
trust
General
trust
A
D
G
J
B
E
H
K
C
F
I
L
items by creating consistent terminology and removing any specic context (e.g., references to a third party logistics (3PL) provider, trucking rm, etc.). To obtain consistency, when items
referred to the respondents rm, we used the term my company, and when items referred to the other rm (e.g., the
respondent rms supplier), we used the term my partner rm.
In addition, we used consistent grammar whereby if the item was
discussing one rm, it was used (i.e., rather than they) and
vice versa. As an example of removing specic context, if an
item indicated, I trust the 3PL, the item was modied to I
trust the partner rm.
As part of this assessment, we also wanted to understand
issues with respect to measurement validity. We have focused
predominantly on validity issues that would be addressed as part
of early scale/survey development and measurement purication
(e.g., content validity, face validity, substantive validity). As
such, evaluating more advanced statistical validity (criterionrelated validity, convergent validity, discriminative validity) associated with the trust measures, while certainly important, is not
addressed in this article.
Our rationale for the focus on a restricted examination of
validity is best summarized with a quote from Dunn et al.
(1994): If content validity does not exist, then there is no reason
to proceed with the analysis because the desired construct is not
being properly represented by the group of items (p. 157).
Churchills (1979, 64) analogy of GIGOgarbage in, garbage
out also applies such that developing measures should start with
specifying and understanding the constructs domain, and, then,
developing items (or searching for existing items) that capture
the domain. In other words, without establishing good measures
at the start, construct validity will be difcult to obtain and/or
reliability will be challenging due to measurement error.
Content validity involves various considerations, and exists
when the construct is properly reected by the items as a
group (Dunn et al. 1994, 157). For example, face validity is
said to exist if the sample is appropriate (e.g., the correct domain
of study) and the survey items appear to respondents as valid
measures of the concept (Mosier 1947; Churchill 1979; Hair
et al. 1998, 117). Although face validity is often considered one
of the weakest measures of validity, face validity can be
enhanced by using experts to review and judge items as well as
by conducting multiple pretests across different populations (Hair
et al. 1998, 117).
Anderson and Gerbing (1991) discuss the concept of substantive validity and illustrated how pretesting techniques (e.g., itemsort task) could be used to rene and eliminate items prior to
survey nalization, which, in turn, should lead to stronger construct validity during empirical testing. While content validity
Conceptualizations of Trust
focuses more on how well the overall scale or items relate to the
construct, substantive validity examines each individual item for
its relevancy to the construct (Anderson and Gerbing 1991; Dunn
et al. 1994).
Item-sort task procedures, as discussed by Anderson and Gerbing (1991), relate closely with Q-sort procedures, and also
incorporate the idea of using a small, expert sample to review
and judge items as proposed by Hair et al. (1998) as well as
other classication schemes (as summarized in Hinkin 1995). Qsorting was invented by British physicist/psychologist William
Stephensen in 1935 as a means of extracting subjective opinion
and evolved from factor analytic theory (Brown 1997).
The Q-sort procedure is used to separate items in a multidimensional construct according to their specic domain (Zait and
Bertea 2011). It can be utilized on an exploratory basis or a conrmatory basis. Exploratory Q-sorting means that respondents
are given the items and asked to group and identify category
labels for each group of items, whereas conrmatory Q-sorting is
used when categories are already labeled and respondents are
asked to classify each item in one category (Zait and Bertea
2011). Conrmatory Q-sorting was used in the current research
project and the input provided by our Q-sort panel helped to
identify items that do not discriminate well between our constructs of interest.
A Q-sort was performed on the 119 unique items. First,
we wanted to determine item consistency with respect to the
level of analysis for each item (e.g., individual vs. organizational
trust). As such, Q-sort participants were asked to examine each
of the 119 items and determine whether the item was focused on
trust at one of the following levels: (1) individual to individual
(e.g., I trust him/her or he/she trusts me); (2) individual to organizational level (e.g., I trust this rm or he/she trusts my company); (3) one organization to one other organization (e.g., my
company trusts this rm or this rm trusts my company); (4) one
organization to many organizations (e.g., my company trusts
multiple rms); or (5) none of the previous choices (e.g., I am
not sure which level is appropriate). Consistency in participant
agreement on the level of analysis was then assessed.
Second, we examined item consistency with respect to how
trust was being conceptualized. As such, Q-sort participants were
asked to examine each item and determine whether the item was
focused on gauging: (1) character-based trust (i.e., dened as
examining the qualitative characteristics of behavior in strategic
philosophies and cultures such as integrity, true motives, consistency and predictability of behavior, openness, and discreteness);
(2) competence-based trust (i.e., dened as examining the specic operating behaviors and day-to-day performance such as
specic competence, knowledge, skills, ability to perform well,
area of expertise, and good judgment); (3) a general level of trust
(e.g., trust is not clearly specied as a specic type); or (4) none
of the previous options (e.g., I am not sure which type of trust is
being referred to in this question).2 Consistency in participant
agreement on the conceptualization was then assessed.
A total of 31 people participated in the Q-sort process. Participants were selected from three different populations to gain
2
121
disciplinary and industry perspectives: 17 participants were academicians familiar with the logistics/SCM literature; six participants were academicians familiar with the marketing/
management literature; and eight participants provided an industry/practitioner perspective.
We assessed consistency from two different perspectives. First,
for each item, we calculated the proportion of substantive agreement, a raw percentage of respondents that assign an item to the
intended category or construct (Anderson and Gerbing 1991). As
we were examining existing items rather than developing the
items and using the Q-sort as a pretest procedure, we did not
identify intended categories a priori. Instead, we report the category that the largest percentage of respondents deemed each item
belonged to, understanding that our categorization (Table 1) may
not be totally representative of the original domain of study for
the 42 articles. For example, if 28 of the 31 participants indicated that an item was examining individual-to-individual trust,
the proportion of substantive agreement was 90% for that item
(28/31). Table 2 shows the summary breakdown for both the
level and type of trust with respect to substantive agreement for
the 119 unique items.
Although the portion of substantive agreement suggests an element of construct validity by illustrating the extent to which an
item is categorized to represent its intended construct, it does not
illustrate the extent to which an item may also be related to a different construct (Anderson and Gerbing 1991). As such, we also
calculated the substantive-validity coefcient for each item, which
examines the extent to which respondents assign an item to the
intended construct more frequently than to a different construct
(Anderson and Gerbing 1991). The substantive-validity coefcient
is calculated by taking the number of respondents that correctly
classify the item to the a priori construct and subtracting the highest number of respondents who assigned the same item to a different construct, and, then, dividing that number by the total number
of respondents (Anderson and Gerbing 1991, 734). Again, we do
not have a priori assignments. As such, for each item, we examined the rst and second category that the largest percentage of
respondents classied for each item. For example, if 22 of the 31
participants indicated that an item was examining individual-toindividual trust, and six participants indicated that the item was
examining individual-to-organizational trust, and three participants
were not sure, the substantive-validity coefcient was .52 for that
item ([22 6]/31). Table 3 shows the summary breakdown for
both the level and type of trust with respect to the substantivevalidity coefcient for the 119 unique items.
Table 2: Q-Sort: Proportion of substantive agreement
Percent agreement
among respondents
100%
9099%
8089%
7079%
6069%
Under 60%
Total items
Level of trust
Type of trust
2
36
18
17
12
34
119
2
22
24
14
22
35
119
122
J. M. Whipple et al.
Level of trust
Type of trust
2
23
15
11
7
15
46
119
2
11
20
12
9
8
57
119
1.0
.90.99
.80.89
.70.79
.60.69
.50.59
Under .50
Total items
trust did not appear until 1995. The vast majority (83%) of the
articles that examine trust have been published in the last
10 years. The results illustrate an increasing interest in examining trust in the logistics/SCM literature. Figure 1 illustrates the
frequency of trust articles published over the 20 years of research
included in this article.
As shown in Figure 2, the majority of the articles (71%) were
published in three journals: Journal of Business Logistics, Journal of Operations Management, and Journal of Supply Chain
Management. Of the full set of articles, the majority (78%) used
existing scales, at least to some extent. As shown in Figure 3,
26% of the articles adopted the full set of items from the cited
source, 31% used a partial set of items from the cited course,
14% used a partial set of items from two or more sources. For
articles using new scales, 7% mixed newly developed scales with
items from a published source while 22% of the articles created
completely new items. Of the nine articles that used new items
and/or new scales, only 5 (55%) provided evidence of, at least,
basic construct validity testing (e.g., Cronbach alpha).
Of the 31 articles that included at least one trust-oriented item
from an existing publication, we assessed the discipline from
which the original source resided. Note: we found numerous
incidents of articles citing a source article only to nd that the
source article was not the original source for the item. It was
also not uncommon for the original source to be at least one
source removed, as authors cited sources of measures that were,
in fact, not the original sources; rather, the source had adopted
original measures from yet a different source article. Figure 4
illustrates that the majority of the source articles were from the
marketing discipline (53%), followed by the organization behavior/management literature (21%). Only two sources originated
within the logistics/SCM literature, and neither source was a
referred journal publication.
Table 4 provides a list of the rst-level source articles. The
table illustrates the number of times the source scales were replicated (i.e., the number of times this particular source article was
used in the 42 logistics/SCM articles examined within this article). Replication does not necessarily mean the full construct
and/or all construct items were used in the subsequent study;
rather, it refers to the use of these measures in some form in
7
6
3%
5
14%
4
45%
1993-1997
1998-2002
2003-2007
38%
1
2011
2009
2007
2003
2005
2001
1999
1997
1995
1993
0
N = 42 ar cles
2008-2012
Conceptualizations of Trust
123
IJLM
5%
TJ
7%
TRE
7%
JSCM
22%
IJPDLM
10%
JBL
25%
JOM
24%
N = 42 ar cles
Figure 3: Construct development summary.
New Items
22%
Full Source
26%
Par al Source
with New Items
7%
Mix of
Sources
14%
Par al Source
31%
N = 42 ar cles
Psychology/Soci
ology
5%
General
Business
5%
Organiza on
Behavior/Manag
ement
21%
Marke ng
53%
IT
5%
SCM
Related
11%
N = 19
Number of
replications
10
6
4
3
2
2
1
1
1
1
1
1
1
1
1
1
1
1
124
J. M. Whipple et al.
Unidimensional
with Specic
Subfactors
2%
Other
19%
Mul dimensional
7%
Agreement on
Level (N=73)
Agreement on
Type (N=62)
Unidimensional
72%
Agreement
Intersec on (N=38)
N = 42
being conceptualized as a construct with subfactorial design.
There was one article that considered trust as a unidimensional
construct made up of specic subfactors (i.e., benevolence and
credibility).
The remaining eight articles were coded as other with
respect to construct dimensionality. In ve of these articles, trust
was measured as a subfactor for a higher order factor. However,
the higher order factor was not trustrather, higher order factors
were constructs, such as relationship magnitude, supply chain
orientation, relationship orientation, and customer partnering
behavior. In the remaining three other articles, trust was a single-item measure used in a plethora of other measures without
construct development.
Table 5 summarizes how each article conceptualized trust as a
construct. As shown in the table, the vast majority of the articles
indicated the construct being measured was trust. In some
articles, two different trust-related constructs were measured,
such as benevolence and partners honesty or calculative trust
and affective trust.
After examining the data to assess the general status of trust
as it has been used in the logistics/SCM literature, the next step
was to analyze the Q-sort data to gain an understanding of elements of content and substantive validity. As indicted in Table 2,
73 items met or exceeded the 70% cutoff point for agreement for
the level of trust, representing 61% of the total 119 items. Of
these 73 items, 52 (71.2%) were categorized as focused on trust
between one organization and one other organization. The
remaining items were categorized as follows: 11 items (15.1%)
were trust between an individual and an organization; six items
(8.2%) were trust between one organization and many organizations; and four items (5.5%) were trust between individuals.
Table 2 also shows that 62 items met or exceeded the 70%
cutoff point for agreement for the type of trust, representing 52%
of the total 119 items. Of these 62 items, 54 of the items
(85.5%) were categorized as character-based trust. The remaining
nine items (14.5%) were categorized as competence-based trust.
Figure 6 depicts the overlap between items that achieved consensus upon level of trust and type of trust. A total of 38 items
had the minimum (70%) substantive agreement on both construct
aspects. As shown in Table 6, the vast majority of these 38 items
were categorized by participants as being character-based trust at
the interorganizational level (one organization to one other organization). Table 7 provides the actual questionnaire items as well
as where each of these 38 items falls within our proposed framework (Table 1).
In addition, we catalogued the items that have the least clarity
with respect to agreement on the level and/or type of trust. Items,
where seven or more Q-sort participants (roughly 25% of the
Q-sort sample) indicated that they did not know what level or
type of trust the item was intended to measure, were deemed to
lack clarity within our framework.3 Furthermore, any items,
which had a substantive-validity coefcient of zero, were also
considered to be poorly specied, as Q-sort participants were
split 50/50 in their assessment of the item as being in one of two
categories (e.g., half of the participants felt the item related to
general trust and half of the participants were unsure which category the item belonged to). Table 8 lists the specic items with
the least agreement.
Frequency of usage
Trust
Benevolence
Partners honesty
Dependability
Objective credibility
Calculative trust
Affective trust
38
4
2
1
1
1
1
DISCUSSION
Logistics and SCM research has become progressively more
sophisticated since the Journal of Business Logistics published
3
Conceptualizations of Trust
125
Type/level
Character-based
trust
Competence-based
trust
General
trust
Total
2
8
23
2
35
0
0
3
0
3
0
0
0
0
0
2
8
26
2
38
Questionnaire item
This person is trustworthy
This person does not make false promises
I feel that this partner rm negotiates honestly
I feel that this partner rm does not try to get
out of commitments
I feel that this partner rm negotiates joint
expectations fairly
I think that the partner rm tells the truth in
business dealings
I feel that the partner rm dealt with me honestly
I feel that the partner rm succeeds by
stepping on other people
I think that the partner rm took advantage
of my problems
I feel that the partner rm takes advantage of
people who are vulnerable
We can count on the partner rm to consider
how its decisions and actions will affect us
When we share our problems with the partner
rm, we are condent that it will be
understanding
In our relationship, the partner rm is sincere
in its promises
In our relationship, the partner rm treats my
company fairly and justly
If the partner rm gives us some advice, we are
certain it is its honest opinion
Regarding problems, this partner rm is always
honest with us
We trust this partner rm keeps our best interests
in mind
This partner rm embodies the values we hold to
be most important
This partner rm deals with us in a straightforward
manner that reveals its true motives and
desired outcomes
My company is accustomed to keeping the best
interests of the partner rm in mind
Table 7: (Continued)
Code from
Table 1
A
A
D
D
D
D
D
D
D
D
G
G
G
G
G
G
G
G
G
G
Continued.
Questionnaire item
The partner rm is fair in its dealing with your
company
We know this partner rm is sincere with us
This partner rm never uses opportunities that
arise to prot at our expense
This partner rm is interested in our companys
well-being and not just its own
This partner rm is truthful in its dealings
with our company
Our company would characterize this partner
rm as being honest
Our company feels that this partner rm cares
about what happens to us
This partner rm is always frank and truthful
in its dealing with us
We believe in this partner rm because it is sincere
This partner rm has always been evenhanded in
its negotiation with us
Even when the partner rm gives us a rather
unlikely explanation, we are
condent that it is telling the truth
The partner rm usually keeps the promises
that it makes to our company
When making important decisions, it is customarily
expected that my company consider the partner
rms welfare alongside its own
The partner rm correctly carries out tasks that we
cannot directly control
This partner rm is competent and effective in its
interactions with our company
It is necessary for us to be condent about the
competence of this partner rm
Our company and these rms are very honest in
dealing with each other
Please describe the degree of trust that exists
between your company and the partner rms:
The extent to which these rms are well known
for their fair dealing
Code from
Table 1
G
G
G
G
G
G
G
G
G
G
G
G
G
H
H
H
J
J
126
J. M. Whipple et al.
Type of trust
Please describe the nature of this partner rms relationship
governance structure (21)
My company expects that all kinds of procedures in the relationship
with the partner rm will become self-evident (10 and SVC)
Whoever is at fault, problems need to be solved together (10)
It would be illogical for the partner rm to abuse my companys
trust (10)
My company will be able to understand the partner rm well (10)
Both parties watch the others protability (9)
The salesperson did everything possible for our company (8)
My company is able to understand the partner rms positive
behaviors (8)
Both parties are willing to make mutual adaptations (7)
Please describe the level of trust toward this partner rm (7)
This salespersons rm has a poor reputation (7)
Notes: *After each item, there is a number in parentheses. This indicates how many of the 31 Q-sort participants did not know how to categorize this
item. Also, for any items where the substantive-validity coefcient was zero, this is indicated as SVC in the parentheses. Italics indicate that the item is
on both the level of trust and the type of trust lists.
its rst issue some 35 years ago. Although there have been
many important contributions to theory and practice, many
opportunities still exist for improvements to the eld. We focus
on one such opportunitybetter scale development. While we
selected trust for examination, it is likely that virtually all of our
elds constructs/research efforts could benet from improved
measures. Developing better scales has the potential for signicantly improving our chance of conducting meaningful research
and theory advancement. Researchers constantly ght the battle
to secure adequate responses and good data. With better, more
precise scales that are user friendly, perhaps, we will enjoy better response rates and lessen participant fatigue.
The lack of replication of trust measures to date suggests that
researchers in the eld are unsatised with the existing measures.
Or, as suggested by McEvily and Tortoriello (2011) when dis-
Conceptualizations of Trust
127
128
J. M. Whipple et al.
Conceptualizations of Trust
129
130
J. M. Whipple et al.
Tyler, T.R. 1990. Why People Obey the Law. New Haven, CT:
Yale University Press.
Whipple, J.M., and Frankel, R. 2000. Strategic Alliance
Success Factors. Journal of Supply Chain Management 36
(3):2128.
Williamson, O.E. 1993. Calculativeness, Trust, and Economic
Organization. Journal of Law and Economics 30:131145.
Zaheer, A., McEvily, B., and Perrone, V. 1998. Does Trust
Matter? Exploring the Effects of Interorganizational and
Interpersonal Trust on Performance. Organization Science 9
(2):14159.
Zait, A., and Bertea, P.E. 2011. Methods for Testing
Discriminant Validity. Management and Marketing 9
(2):21724.
Zucker, L.G. 1986. Production of Trust: Institutional Sources of
Economic Structure, 18401920. Research in Organizational
Behavior 8:53111.
SHORT BIOGRAPHIES
Judith M. Whipple (PhD Michigan State University) is an
Associate Professor in the Department of Supply Chain Management at Michigan State University. Her research interests include
collaboration, risk management, and supply chain organizational
design. Judy has published work in various logistics/supply chain
management journals.
Stanley E. Grifs (PhD The Ohio State University) is an
Associate Professor in the Department of Supply Chain Management at Michigan State University. His research interests include
supply chain disruptions, customer valuation of logistics, and
network design. He has published work in a variety of logistics,
supply chain, and related journals.
Patricia J. Daugherty (PhD Michigan State University) holds
the Bowersox Chair in Logistics and Supply Chain Management
in the Department of Supply Chain Management at Michigan
State University. She is a former editor of the Journal of Business Logistics and has published widely in logistics and supply
chain journals.
s supply chains spread toward emerging economies, Western buying rms frequently face the question of whether they should commit
resources to develop their suppliers in these regions in terms of environmental issues. Supplier development researchers have just begun
to consider environmental aspects, and thus far, the peer-reviewed literature has remained primarily qualitative, and often descriptive. Largescale empirical evidence indicating the antecedents and benets of environmental supplier development for a buying rm is still scarce.
Addressing this gap, we use stakeholder theory to complement and extend the work of Ehrgott et al. (2011) and investigate how pressures from
customers, governments, and employees act as antecedents to environmental supplier development. Furthermore, we build on the resource-based
view to examine how supplier capabilities, buying rm environmental reputation, and organizational learning in the buying rm can result from
such supplier development initiatives. We test the resulting model with a sample of 244 corporate procurement executives from the United
States and Germany.
Keywords: environmental sustainability; supplier development; emerging markets; international purchasing; stakeholder theory; resource-based
view
INTRODUCTION
As supply chains spread toward emerging economies (Kusaba
et al. 2011; Geref and Lee 2012), the debate about the responsibility of Western buying rms to educate their suppliers with
regard to environmental issues has become increasingly relevant
to supply chain and sustainability research. Suppliers from many
parts of Asia, Eastern Europe, and Latin America often lack the
necessary managerial and technological capacities, as well as the
awareness to address the environmental issues inherent in their
businesses (Child and Tsai 2005; Zhu et al. 2011). Thus, buying
rms in developed nations that source from these regions face
the question of whether to commit resources to environmental
concerns beyond their own organizational boundaries, toward
their upstream supply chains (Pagell and Wu 2009; Bai and Sarkis 2010). The supplier development literature has only recently
started to recognize such environmental efforts as part of rms
supplier development initiatives. This recent literature has
focused on providing a descriptive overview, formal modeling,
or case-based investigation of issues related to environmental
supplier development (Pagell and Wu 2009; Bai and Sarkis
2010; Reuter et al. 2010; Fu et al. 2012). While Ehrgott et al.
(2011) examine the inuence of key stakeholders to socially sustainable supplier selection, we note a dearth of large-scale
empirical studies that examine the stakeholder antecedents to
buying rms engagement in environmental supplier development
efforts, and the precise benets the buying rm can reap through
such engagement. Notable exceptions that have made initial
headway into the phenomenon are Eltayeb et al. (2010) and
Vachon and Klassen (2008). We add to this evolving stream by
Corresponding author:
Matthias Ehrgott, WHUOtto Beisheim School of Management,
Burgplatz 2, 56179 Vallendar, Germany; E-mail: matthias.ehrgott@
whu.edu
investigating two research questions: (1) Which stakeholder pressures act as antecedents for Western buying rms to engage in
developing the environmental capabilities of their emerging economy suppliers? and (2) Which tangible benets can rms capture
through such environmental supplier development efforts?
Our denition of environmental supplier development
encompasses all activities through which the buying rm helps
its suppliers reduce their negative environmental impact. Such
activities typically include environmental supplier education
(e.g., training that reduces emissions through better ltering and
waste treatment, as well as improved efciency in production
technology), environmental on-site supplier support (e.g., provision of on-site technical assistance to redesign existing processes), and joint environmentally oriented business projects
(e.g., collaboration to jointly develop green innovations and technologies) (Carter and Carter 1998; Min and Galle 2001; Eltayeb
et al. 2010). Also, such activities typically involve several of the
buying rms functions (including purchasing, research and
development [R&D], production, human resources, and logistics)
to collaborate in a cross-functional team (Schaefer and Kosansky
2008; Paulraj 2011). We denote rms environmental efforts in
general (i.e., beyond environmental supplier development) using
the term environmental sustainability and efforts including both
environmental and social sustainability aspects by the broader
term sustainability.
The theoretical lenses through which we examine our research
questions are stakeholder theory (Freeman 1984; Frooman 1999)
and the resource-based view (RBV) of the rm (Wernerfelt
1984; Barney 1991)concepts that scholars frequently draw on
for investigating sustainability efforts (Clarkson 1995; Hart 1995;
Sharma and Vredenburg 1998; Aragon-Correa and Sharma 2003;
Garriga and Mel!e 2004; Ehrgott et al. 2011).
Stakeholder theory posits that rms must consider a broader set
of stakeholder interestsbeyond nancial rm performanceto
maintain their business activities in the long run (Freeman 1984).
These interests increasingly include the preservation of the natural
132
M. Ehrgott et al.
kets list, resulting in the inclusion of 50 relevant sourcing countries. (See the Methodology section and Appendix 1 for further
details.)
The remainder of this article is organized as follows. Next, we
provide a brief review of the literature on environmental supplier
development. We then develop the hypotheses that determine our
theoretical framework, using an antecedent-environmental supplier
development-outcome structure and building on prior research in
the eld. Subsequently, we describe the studys methodology. In
the nal sections, we present the empirical ndings, discuss implications and limitations, and identify directions for future research
in the area.
ENVIRONMENTAL SUPPLIER DEVELOPMENT
LITERATURE AND CONCEPT
During the past two decades, the literature on supplier development has evolved signicantly (Hartley and Jones 1997; Krause
and Ellram 1997; Krause et al. 1998; S!anchez-Rodriguez et al.
2005; Wagner 2006). Leenders (1966) describes supplier development as a buying rms efforts to improve its suppliers performance. More recently, scholars in the area have started to
argue for a more ne-grained perspective on supplier developmentone that accounts for the conceivably very different goals
behind such initiatives (Krause 1999; Wagner and Krause 2009).
In this respect, Krause (1999) nds improvements in quality,
delivery, cost structure, new technology adoption, nancial
health, and product design capability as frequent underlying
objectives. Krause and Handeld (1999) add to this list the
improvement of communication and alignment of procedures in
the buyersupplier relationship, which aim at lowering future
coordination costs.
Depending on the priority of these goals for the buying rm, the
magnitude and type of interventions with the suppliers processes
and operations as part of supplier development programs differ
(Wagner and Krause 2009). For example, delivery-oriented
supplier development might include changes in the suppliers
logistics and production planning routines. Product design-oriented
supplier development, by contrast, might focus more on changes in
R&D processes, the product input factor portfolio, and production/
assembly equipment and processes.
The notion that the development of suppliers might also serve
environmental goals has only recently entered the supply chain
literature. In this vein, Bai and Sarkis (2010) note that investigation into [environmental] supplier development programs is
virtually non-existent (p. 1200). One of the few established
terms in the eld was coined by Vachon and Klassen (2008).
They refer to joint ecological efforts between buying rms and
suppliers as environmental collaboration, which typically
includes environmental supplier training (to spur initiatives such
as emissions reduction, advanced waste treatment, and resource
efciency), on-site intervention with suppliers production processes and operations to improve their environmental footprint,
and joint projects to develop green innovations (Carter and Carter 1998; Min and Galle 2001; Eltayeb et al. 2010; Azadegan
2011). Several major corporate projects targeted at environmental
supplier development have recently been made public. BASF,
the multinational chemicals conglomerate, has launched a joint
133
134
supplier development can reduce the risk of spills and other environmentally harmful incidents at the supplier sites (Lee and Klassen 2008) that the buying rms customers would consequently
judge negatively. Together, these considerations suggest that environmental supplier development is a suitable response to environmental customer pressures. Hence:
H1: Customer environmental pressures are positively
related to environmental supplier development.
Government environmental pressures
Government pressures refer to the strength with which a rms
home country government pushes rms to undertake greater
environmental efforts. We opt for this home country perspective
(Ehrgott et al. 2011) because environmental regulation in the
United States and Europe has seen a strong rise in stringency
(Emmelhainz and Adams 1999; Goldsby and Stank 2000; Mollenkopf et al. 2005) and ranks among the most advanced with
regard to covering rms upstream supply chains (Linton et al.
2007). Our conceptualization incorporates not just the level of
current environmental regulatory pressures but also expectations
about future pressures.
In terms of the rms environmental supplier development
efforts, expectations about future, heightened regulation are particularly likely to inuence rm conduct. Firms that expect regulators to set higher environmental standards in the future would
tend to engage in supplier development in the present to prepare
for the changes, thus ensuring that their supply base meets the
anticipated requirements. In doing so, the buying rm reduces
the risk of high readjustment costs (e.g., supplier switching costs)
if tighter environmental regulations are implemented (Barnett and
King 2008; Ehrgott et al. 2011). Environmental supplier development in itself might not be the most obvious response to environmental government pressures because it seldom addresses the
entire supplier base (thus not necessarily leading to consistent
standards) and it takes time to implement (thus not effectively
addressing regulation that requires immediate adherence). Still,
looking at prospective future tightening of regulation, environmental supplier development allows the buying rm to keep
business relationships with suppliers that are not or will not be
able to reach the required environmental standards by themselves, but who are generally good business partners. Hence:
H2: Government environmental pressures are positively
related to environmental supplier development.
Middle management environmental pressures
We conceptualize environmental pressure from employees by
focusing on the inuence of one specic group of employees:
middle managers in the purchasing function. Here, we build on
the work of Drumwright (1994), who nds that most [originators of environmental initiatives] were well-respected middle
managers who had a claim to a hearing through some type of
power (p. 4). We dene environmental pressure from middle
management as the extent to which purchasing middle managers,
in self-motivated and proactive ways, push their employer to
improve the environmental standards in its supplier base. Partici-
M. Ehrgott et al.
strategic capabilities, including their technological and managerial skills (Vachon and Klassen 2008), from which the buying
rm ultimately benets as well. Also, environmental supplier
development can be perceived as an extraordinary environmental
commitment by the buying rm, and we expect such a commitment to positively affect the buying rms environmental
reputation. Finally, we argue that environmental supplier development can be a platform for substantial organizational learning,
not only on the supplier side but also, particularly, for the buying
rm itself.
Suppliers strategic capabilities
Suppliers strategic capabilities center on the longer term organizational characteristics of the rms suppliers. Management quality (Stump and Heide 1996), nancial stability (La Londe and
Masters 1994), and technological capabilities (Monczka et al.
1995) can be seen as key reections of these characteristics (Ehrgott et al. 2011). For the buying rm, understanding suppliers
strategic capabilities is critical because these capabilities determine the suppliers capacity to innovate and remain competitive
and thus to participate in a stable and longer term business relationship. Managers interviewed during our survey pretest
described different ways in which buying rms conrm their
suppliers strategic capabilities. For example, in the regular supplier selection and monitoring processes, buying rms typically
assess the development of key revenue and protability performance indicators over the past several years, compliance with
the latest ISO certications, and reports of technological competencies and capacities (e.g., through supplier self-reporting instruments) as proxies for such capabilities.
We expect that investments in the development of suppliers
environmental standards positively affect the general strategic
capabilities of these suppliers. Contrary to some types of supplier
development (e.g., logistics-centered development projects that
do not necessarily intervene in a broad way with the suppliers
production technology), upgrading a suppliers environmental
performance often entails broader technological intervention.
It typically requires helping the supplier use technologies and
production processes that are more advanced and that specically
allow for improvements such as lower emissions footprints, more
efcient energy use, and/or better waste management (Klassen
and Vachon 2003). For example, BASFs eco-efciency development project with emerging economy suppliers involved a systematic modernization of the suppliers production systems; it
started with an analysis of the suppliers current status using
BASFs proprietary diagnostic tools and included consulting on
process and technological improvements throughout the suppliers systems by BASFs production experts (UNIDO 2004). Both
the outlined reasoning and the BASF example align with the
contentions of the RBV. Specically, Hart (1995) and Waddock
and Graves (1997) nd that adherence to high environmental
standards often requires the development of superior management and technology resources, including advanced manufacturing technologies and efcient and tightly controlled processes.
These resources allow not only for better environmental performance but also for higher product sophistication and innovation
capabilities. Hence:
135
136
METHODOLOGY
Sampling and data collection
To collect the studys empirical data, we used a mail questionnaire sent to a random sample of 1,500 U.S. companies and
1,000 German companies from the manufacturing, construction,
M. Ehrgott et al.
This sample resulted from the same data collection effort that
is the basis of Ehrgott et al. (2011).
137
138
M. Ehrgott et al.
1
.468
.605
.537
.196
.253
.057
.283
.103
.133
.030
1
.365
.471
.107
1
.172
.039
1
.050
1
.550
.229
.084
.108
.024
Efforts
Outcomes
R2 : .01
Customer
environmental
pressures
H1
Suppliers
strategic
capabilities
H4
R2: .29
Government
environmental
pressures
Environmental
supplier
development
H2
R2 : .22
.37***
.58***
Middle mgmt
environmental
pressures
Buying firm
environmental
reputation
H5
.47***
H3
H6
R2 : .13
Organizational
learning
in supplier mgmt
Model fit: X2 /df = 2.159 CFI = .937 TLI = .927 GFI = .856 AGFI = .819 NFI = .889 RMSEA = .069
*: p<0.05; **: p<0.01; ***: p<0.001
Significant relationship
n= 244
Insignificant relationship
Note: The antecedent constructs are dened based on Ehrgott et al. (2011) but use separate survey items which operationalize these pressures as environmental, rather than social, stakeholder pressures. The suppliers strategic capabilities and the organizational learning in
supplier management constructs use the same data used by Ehrgott et al. (2011).
sured as a percentage of total purchasing volume in U.S. dollars/Euros for the U.S. subsample and the German subsample,
respectively.
When environmental sustainability is part of a rms overall
business strategy, middle managers might be encouraged to more
explicitly push for environmental projects. For the same strategic
reasons, however, the rm might also engage in environmental
supplier development (Carter and Jennings 2004), potentially
leading to endogeneity issues. We therefore control for the inuence of environmental rm strategy using the item, Our company makes environmental sustainability an integral part of its
business strategy, on a 7-point Likert-type scale.
Similarly, a people-oriented organizational culture has been
found to affect employee initiatives, as well as rms conduct concerning environmental issues (Carter and Jennings 2004). To control for such an inuence, we include a scale based on the work of
Chatman and Jehn (1994) and Carter and Jennings (2004) with the
following items: Our companys organizational culture is strongly
characterized by (a) being people-oriented, (b) fairness, (c) being
supportive, and (d) the desire to be a good corporate citizen.
139
demands for environmental sustainability. Firms might in fact recognize that they can change their supplier portfolios relatively easily at a later point in time, as an immediate response to regulatory
changes (Ehrgott et al. 2011). The need for a proactive stance on
government regulation would thus be lower than expected. That is,
rms would not need to engage in costly environmental supplier
development, but instead would switch to better suppliers when
needed. Such switching is even likely to be relatively neutral in
cost, on average, in emerging economies because Western rms
still tend not to fully integrate with suppliers from these regions,
but rather to source commodities or simpler components from
them. Support for this view comes from Buysse and Verbeke
(2003), nding a positive governmental inuence only on environmentally reactive companies and not on environmentally proactive
rms.
Outcomes
The most striking nding with regard to outcomes of the rms
environmental supplier development is the highly signicant
links between supplier development efforts and both buying rm
environmental reputation (H5) and organizational learning (H6).
Apparently, environmental supplier development helps to build
a positive environmental reputation as it is perceived as an
extraordinary engagement that requires high resource commitments (Bai and Sarkis 2010) and demonstrates the buying rms
ability to manage complex environmental challenges in its
upstream supply chain. Our nding thus supports the work by
Hasseldine et al. (2005), suggesting that even a smaller number
of high-involvement supplier development efforts apparently is
able to augment the rms environmental reputation.
Based on our empirical evidence regarding organizational
learning, environmental supplier development efforts appear to,
in fact, lead to sufciently rich interaction between the purchasing department and suppliers, so as to generate strong learning
effects for the buying rm. Environmental supplier development
projects can require that many different experts from the buying
rm be involved at the supplier site for extended periods of time
(Carter and Carter 1998). As evidenced by the previously
described examples of BASF, Pentland, and Nike (Bethke and
Bluethner 2003; UNIDO 2004), environmental development projects in emerging economies allow for an exchange of thoughts
beyond the usual business context, providing a broader perspective of the suppliers political, legal, and social context in the
emerging region. This rich interaction gives the buying rm the
opportunity to gain deep insight into the business practices of
emerging economy rms and thus to enhance its overall ability
to manage relationships with business partners from these
regions.
In contrast, we did not nd a relationship between environmental supplier development and suppliers strategic capabilities
(H4). One explanation could be that the rm is not able to reach
a sufcient number of suppliers with such development efforts to
measurably affect the average capabilities of its supplier base (as
measured in this study). This thought is supported by the work
of Klassen and Vachon (2003), who do nd a positive effect of
collaborative environmental initiatives (e.g., supplier development initiatives) on supplier capabilities when they look at the
individual suppliers involved in such initiatives rather than at the
140
M. Ehrgott et al.
141
APPENDIX 1
ALPHABETICAL LIST OF EMERGING SUPPLY MARKETS IN SCOPE (AS PRESENTED TO RESPONDENTS)
Eastern Europe
- Albania
- Armenia
- Azerbaijan
Asia
- Georgia
- Hungary
- Kazakhstan
- Romania
- Russia
- Slovakia
- Bangladesh
- China
- India
Latin America
- South Korea
- Sri Lanka
- Taiwan
- Argentina
- Brazil
- Chile
Continued.
142
M. Ehrgott et al.
Appendix 1: (Continued)
Eastern Europe
- Belarus
- Bosnia/
Herzegovina
- Bulgaria
- Croatia
- Czech
Republic
- Estonia
Asia
-
Kyrgyzstan
Latvia
Lithuania
Macedonia
Moldova
Poland
Slovenia
Tajikistan
Turkey
Turkmenistan
- Ukraine
- Uzbekistan
Indonesia
Iran
Malaysia
Pakistan
Philippines
Singapore
Latin America
- Thailand
- Vietnam
Columbia
Ecuador
Mexico
Peru
Trinidad
Venezuela
APPENDIX 2
SAMPLE DEMOGRAPHICS
Characteristic
Number of responses
Average rm size in sales per annum
Average # of employees
% Electronic rms
% Automotive rms
% Chemicals, pharmaceuticals, oil and gas rms
% Mechanical engineering rms
% Consumer goods rms
% Wholesale and retail trading rms
% Other rms
Average import ratio from emerging economies
U.S. subsample
German subsample
107
USD 6,659 M*
17,357
12.1%
13.1%
15.9%
18.7%
19.6%
4.7%
15.9%
23.46%
137
USD 6,734 M*
21,901
6.6%
16.8%
11.7%
19.0%
24.8%
6.6%
14.6%
29.19%
Note: * Converted into U.S. dollars using EUR-USD exchange rate at the time of survey.
APPENDIX 3
QUESTIONNAIRE SCALE ITEMSa,b,c
Customer Environmental Pressures
Our customers
set high environmental standards in their buying decision. (.94)
show strong awareness about environmental issues. (.94)
prefer purchasing from companies with a strong environmental image. (.91)
inform themselves about their suppliers environmental standards before buying from them. (.86)
Composite Reliability: 0.93, Cronbachs Alpha: 0.93, AVE: 0.772
Eliminated item (reason for elimination):
are likely to switch to our competitors if these outperform us in environmental aspects. (high internal correlation)
Government Environmental Pressures
Government/legal regulation in our industry
currently sets strict environmental standards. (.88)
is likely to increase pressure if our industry does not improve environmentally by itself. (.92)
actively pushes for environmental improvement. (.94)
Composite Reliability: 0.90, Cronbachs Alpha: 0.90, AVE: 0.761
Eliminated items (reason for elimination):
Continued.
143
is expected to increase pressure regarding environmental efforts within the next three years. (high internal correlation)
is lobbied by activist groups to increase environmental standards. (high internal correlation)
currently holds my company responsible for the environmental impact of our suppliers. (low factor loading)
Middle Management Environmental Pressures
The middle managers in our purchasing organization
show a personal sense of obligation towards environmental conduct. (.91)
want our company to be perceived as a leader in terms of environmental responsibility (e.g., by the public, customers, media, etc.).
(.91)
speak up if they feel our company can improve environmentally. (.90)
Composite Reliability: 0.89, Cronbachs Alpha: 0.89, AVE: 0.738
Eliminated items (reason for elimination):
have started projects to enhance our companys environmental performance. (high internal correlation)
show initiative to advance environmental causes. (high internal correlation)
Environmental Supplier Development
With regard to our existing portfolio of suppliers from emerging economies, we
advise them on technology that makes their production cleaner. (.89)
make efforts (e.g., joint projects) to show them how they can reduce waste in manufacturing. (.95)
make efforts (e.g., joint projects) to show them how they can use resources more efciently. (.94)
Composite Reliability: 0.92, Cronbachs Alpha: 0.91, AVE: 0.785
Eliminated items (reason for elimination):
encourage them to continuously improve their emission levels (e.g., air or water pollution). (low factor loading)
commit resources to coach them on employing reusable packaging solutions. (high internal correlation)
continuously ask them to commit to waste reduction goals. (high internal correlation)
Suppliers Strategic Capabilities
Averaging over our total purchasing volume from emerging economies, our current suppliers from these regions show
high technical capabilities. (.85)
nancial strength. (.80)
good personnel and management resources. (.80)
high innovation capabilities. (.85)
Composite Reliability: 0.84, Cronbachs Alpha: 0.84, AVE: 0.576
Buying Firm Environmental Reputation
Our company
is perceived as a leader in terms of environmental responsibility by our customers. (.89)
is often referred to as a role model for environmental business practices by the media. (.91)
is seen as an innovator regarding environmental business practices in our industry. (.92)
Composite Reliability: 0.89, Cronbachs Alpha: 0.89, AVE: 0.738
Eliminated items (reason for elimination):
is the preferred provider for customers who care about environmental issues. (high internal correlation)
has won awards for environmental merits. (high internal correlation)
Extent of Organizational Learning in Supplier Management
Interacting with suppliers from emerging economies enabled us to
develop skills to effectively coordinate with suppliers from such regions. (.84)
develop expertise to manage suppliers from such regions. (.90)
gain expertise that we can use across different countries in these regions. (.69)
Composite Reliability: 0.85, Cronbachs Alpha: 0.86, AVE: 0.649
Eliminated items (reason for elimination):
learn more about doing business in these regions. (high internal correlation)
gain insights that are also useful for other functions of our rm. (high internal correlation)
Fit indices overall measurement model: v/df = 1.52; CFI = 0.97; TLI = 0.97; GFI = 0.90; AGFI = 0.87; NFI = 0.93;
RMSEA = 0.046.
Notes: AVE, average variance extracted.
a
Standardized factor loadings in parentheses.
b
All items were measured on a 7-point Likert scale where 1 = strongly disagree and 7 = strongly agree.
c
Respondents were told, This survey is targeted at companies that purchase from emerging economies. (The country list as shown in
Appendix 1 was provided at the end of the questionnaire.) They were instructed: You should only continue if your company buys part
of its total purchasing volume, including direct material, indirect material, and/or machinery & equipment, from emerging economies
(suppliers shipping plants should be located in these regions). Furthermore, they were asked If your company has more than one
business unit, please answer all questions with regard to your business unit only.
144
REFERENCES
Anderson, J.C., and Gerbing, D.W. 1988. Structural Equation
Modeling in Practice: A Review and Recommended TwoStep Approach. Psychological Bulletin 103(3):41123.
Aragon-Correa, J.A., and Sharma, S. 2003. A Contingent
Resource-Based View of Proactive Corporate Environmental Strategy. Academy of Management Review 28(1):71
88.
Arens, P., and Brouthers, K.D. 2001. Key Stakeholder Theory
and State Owned Versus Privatized Firms. Management
International Review 41(4):37794.
Armstrong, J.S., and Overton, T.S. 1977. Estimating
Nonresponse Bias in Mail Surveys. Journal of Marketing
Research 14(3):396402.
Auger, P., Burke, P., Devinney, T.M., and Louviere, J.J. 2003.
What Will Consumers Pay for Social Product Features?
Journal of Business Ethics 42(3):281304.
Azadegan, A. 2011. Beneting From Supplier Operational
Innovativeness: The Inuence of Supplier Evaluations and
Absorptive Capacity. Journal of Supply Chain Management
47(2):4964.
Backhaus, K.B., Stone, B.A., and Heiner, K. 2002. Exploring
the Relationship Between Corporate Social Performance and
Employer Attractiveness. Business and Society 41(3):292
318.
Bai, C., and Sarkis, J. 2010. Green Supplier Development:
Analytical Evaluation Using Rough Set Theory. Journal of
Cleaner Production 18(12):120010.
Barnett, M.L. 2007. Stakeholder Inuence Capacity and the
Variability of Financial Returns to Corporate Social
Responsibility. Academy of Management Review 32(3):794816.
Barnett, M.L., and King, A. 2008. Good Fences Make Good
Neighbors: A Longitudinal Analysis of an Industry SelfRegulatory Institution. Academy of Management Journal 51
(6):115070.
Barney, J. 1991. Firm Resources and Sustained Competitive
Advantage. Journal of Management 17(1):99120.
Barney, J. 2012. Purchasing, Supply Chain Management, and
Sustained Competitive Advantage: The Relevance of
Resource-Based Theory. Journal of Supply Chain
Management 48(2):36.
Bentler, P.M. 1989. EQS Structural Equations Program Manual.
Los Angeles: BMDP Statistical Software.
Bethke, K., and Bluethner, A. 2003. Eco-Efcient Production
for SME Suppliers. Global Compact Policy Dialogue June
12/13 2003, New York.
Browne, M.W., and Cudeck, R. 1989. Single Sample CrossValidation Indices for Covariance Structures. Multivariate
Behavioral Research 24(4):44555.
Buysse, K., and Verbeke, A. 2003. Proactive Environmental
Strategies: A Stakeholder Management Perspective. Strategic
Management Journal 24(5):45370.
Cantor, D.E., Morrow, P.C., and Montabon, F. 2012.
Engagement in Environmental Behaviors Among Supply
Chain Management Employees: An Organizational Support
Theoretical Perspective. Journal of Supply Chain
Management 48(3):3351.
M. Ehrgott et al.
Carter, C.R. 2000. Ethical Issues in International BuyerSupplier Relationships: A Dyadic Examination. Journal of
Operations Management 18(2):191208.
Carter, C.R. 2005. Purchasing Social Responsibility and Firm
Performance: The Key Mediating Roles of Organizational
Learning and Supplier Performance. International Journal of
Physical Distribution & Logistics Management 35(3):17794.
Carter, C.R., and Carter, J.R. 1998. Interorganizational
Determinants of Environmental Purchasing: Initial Evidence
From the Consumer Products Industries. Decision Science 29
(3):65984.
Carter, C.R., Ellram, L.M., and Tate, W. 2007. The Use of
Social Network Analysis in Logistics Research. Journal of
Business Logistics 28(1):13770.
Carter, C.R., and Jennings, M.M. 2004. The Role of Purchasing
in Corporate Social Responsibility: A Structural Equation
Analysis. Journal of Business Logistics 25(1):14586.
Carter, C.R., and Rogers, D. 2008. A Framework of Sustainable
Supply Chain Management: Moving Toward New Theory.
International Journal of Physical Distribution and Logistics
Management 38(5):36087.
Chatman, J.A., and Jehn, K.A. 1994. Assessing the
Relationship
Between
Industry
Characteristics
and
Organizational Culture: How Different Can You Be?
Academy of Management Journal 37(3):52253.
Chen, H., Daugherty, P.J., and Landry, T.D. 2009. Supply
Chain Process Integration: A Theoretical Framework.
Journal of Business Logistics 30(2):2746.
Child, J., and Tsai, T. 2005. The Dynamic Between Firms
Environmental Strategies and Institutional Constraints in
Emerging Economies: Evidence From China and Taiwan.
Journal of Management Studies 42(1):94125.
Christmann, P. 2004. Multinational Companies and the Natural
Environment: Determinants of Global Environmental Policy
Standardization. Academy of Management Journal 47
(5):74760.
Ciliberti, F., Baden, D., and Harwood, I.A. 2009. Insights Into
Supply Chain Pressure on CSR-Practice: A Multi-Case Study
of SMEs in the UK. Operations and Supply Chain
Management 2(3):15466.
Clarkson, M. 1995. A Stakeholder Framework for Analyzing
and Evaluating Corporate Social Performance. Academy of
Management Review 20(1):92117.
Deephouse, D.L., and Heugens, P. 2009. Linking Social Issues
to Organizational Impact: The Role of Infomediaries and the
Infomediary Process. Journal of Business Ethics 86:54153.
De Leeuw, S., and Fransoo, Y. 2009. Drivers of Close Supply
Chain Collaboration: One Size Fits All?. International Journal
of Operations & Production Management 29(7):72039.
Drumwright, M.E. 1994. Socially Responsible Organizational
Buying: Environmental Concern as a Noneconomic Buying
Criterion. Journal of Marketing 58(3):119.
Dutton, J.E., Dukerich, J.M., and Harquail, C.V. 1994.
Organizational Images and Member Identication.
Administrative Science Quarterly 39(2):23963.
Ehrgott, M., Reimann, F., Kaufmann, L., and Carter, C.R. 2011.
Social Sustainability in Selecting Emerging Economy
Suppliers. Journal of Business Ethics 98:99119.
145
146
M. Ehrgott et al.
147
SHORT BIOGRAPHIES
Matthias Ehrgott (PhD WHUOtto Beisheim School of
Management) is an Assistant Professor of International Business
& Supply Management at WHUOtto Beisheim School of
Management. His primary research stream focuses on sustainability issues in the supply chain, and his work has appeared in the
Journal of Business Ethics, the Journal of Purchasing & Supply
Management, and the Journal of International Management,
among others. For his doctoral research on social and environmental sustainability in supplier management, Dr. Ehrgott has
been honored by the Council of Supply Chain Management Professionals (CSCMP) with the Doctoral Dissertation Award. Dr.
Ehrgott is a member of the editorial review boards of the Journal
of Operations Management and the Journal of Supply Chain
Management and has several years of experience in industry.
Felix Reimann (PhD WHUOtto Beisheim School of Management) is an Assistant Professor of International Business at
WHUOtto Beisheim School of Management. His primary
research stream focuses on sustainability issues in emerging
economy contexts, and his work has appeared in the Journal of
Business Ethics, the Journal of Purchasing & Supply Management, and the Journal of International Management, among others. Dr. Reimann is a member of the editorial review board of the
Journal of Supply Chain Management and has several years of
experience in industry, particularly in Asia and Europe.
Lutz Kaufmann (PhD University of Giessen) is a Professor of
International Business & Supply Management and member of the
Supply Chain Management group at WHUOtto Beisheim
School of Management. He is also an Associate Fellow of Said
Business School at Oxford University, UK. His research has been
published in the Journal of Business Logistics, Journal of Operations Management, Journal of Purchasing & Supply Management,
Journal of International Management, Journal of World Business,
Journal of International Marketing, Journal of Business Ethics,
International Journal of Physical Distribution & Logistics Management, and Journal of Supply Chain Management, among others. Dr. Kaufmann is the European editor of the Journal of Supply
Chain Management.
Craig R. Carter (PhD Arizona State University) is a Professor of Supply Chain Management at Arizona State University.
His primary research focus is on sustainable supply chain management, which encompasses ethical issues in buyersupplier
relationships, environmental supply management, diversity sourcing, perceptions of opportunism surrounding electronic reverse
auctions, and the broader integrative concepts of social responsibility and sustainability. He is a member of the editorial review
board of several journals, an associate editor of the Journal of
Business Logistics and Journal of Operations Management, and
the co-editor-in-chief of the Journal of Supply Chain Management. His research has appeared in numerous supply chain management journals, including the Journal of Business Logistics,
Journal of Supply Chain Management, Decision Sciences, and
Journal of Operations Management.
ogistics contribution to corporate performance has increased over recent years, particularly due to supply chain innovations. Opposed to
common innovations focusing on the improvement of product or information ow, supply chain nance (SCF) targets the nancial ow
and allows buying rms and their suppliers to improve working capital and reduce costs. However, the adoption process of SCF is complex
and rather unexplored in academia. This article provides an early step in building knowledge about SCF and in particular how rms adopt SCF,
why they adopt differently, and what role suppliers play in the adoption process. The objective was therefore to close the gap between our
knowledge on product and information ow oriented innovations and nancial ow innovations along the supply chain, namely SCF. For this
explorative research, we opted for an inductive multiple case study approach with six European rms. Based on our ndings, four sets of propositions are posited and an extended SCF adoption framework is proposed revolving around the interrelated adoption processes of buying rms
and their corresponding supplier bases.
Keywords: supply chain nance; innovation adoption; upstream innovation; case studies
INTRODUCTION
It is widely acknowledged that superior logistics management is
a crucial driver of rm performance (Ellram 1991; Bowersox and
Closs 1996; Mentzer et al. 2004; Fugate et al. 2010). Several
innovations such as bar codes, radio frequency identication,
cross-docking, and just-in-time delivery helped to grow the strategic impact of logistics management. Each of these innovations
supported specic rms to outcompete competitors, with prominent examples such as Walmart, Zara, Amazon, Toyota, and Dell
(Chopra and Meindl 2012). The scope was traditionally limited
to managing physical inventory and information ows, whereas
paying less attention to innovations in the third logistical ow,
the nancial ow of supply chains.
Lately, we have seen several rms tapping into the eld of
supply chain nance (SCF) as practitioner reports show (Aberdeen Group 2006, 2007; Demica 2007). In such reports, it is
claimed that even one of seven rms actively uses SCF (Aberdeen Group 2007), as the need to harmonize nancial and physical ows in European supply chains has been substantial, even
already before the nancial crisis (Castill!
on and Petit 2008). SCF
provides a pathway not only out of short-term liquidity dilemmas
but also toward a reduction in the long-term nancial burden in
the supply chain represented, for example, by the total amount of
necessary liquidity in a supply chain. The necessary liquidity is
lower with a coordinated nancial ow across the supply chain
than in the uncoordinated case (Protopappa-Sieke and Seifert
2010), especially leading to high savings when buyers and suppliers have different credit ratings (Pfohl and Gomm 2009).
Corresponding author:
David A. Wuttke, Institute for Supply Chain Management, Procurement and Logistics (ISCM), EBS University of Business and Law,
EBS Business School, Konrad-Adenauer-Ring 15, Wiesbaden
65187, Germany; E-mail: David.Wuttke@ebs.edu
Only recently, scholars also started to emphasize the importance of managing nancial ows along supply chains (Bowersox and Closs 1996; Mentzer et al. 2001; Hofmann and Kotzab
2010; Gupta and Dutta 2011) and to address research topics with
adjacent focus, for example, Protopappa-Sieke and Seifert (2010)
on the interrelation of operational and nancial performance
measures in inventory control. Likewise, Hofmann (2009) studies
inventory nancing from a logistics service provider perspective
and Pfohl and Gomm (2009) and Gomm (2010) review and conceptualize different approaches of nancing supply chains. A conceptual approach is also provided by Hofmann and Kotzab
(2010) who study collaborative working capital management and
particularly cash management in supply chains. However, empirical knowledge about this new phenomenon of SCF is nascent,
which can be explained by the mere fact that the SCF innovation
only recently emerged and empirical research can only analyze
existing practices and phenomena.
In practice, the innovation of SCF is considered to be an
established structure founded on an agreement between a buying
rm with its bank, stating that any supplier whose invoice has
been released by this buying rm can obtain a credit from the
bank for the period of the payment terms against the buying
rms credit rating (e.g., Demica 2007). This process is often
automated through an electronic platform providing all involved
parties with real-time visibility into the relevant nancial transactions. Particularly in Europe, SCF offers some innovative aspects
opposed to related practices such as reverse factoring, which are
intensively used in emerging economies (Klapper 2006).
Figure 1 illustrates the differences of a transaction without and
a transaction including SCF indicating also the novelty of the
process. This gure depicts the basic mechanism of SCF as we
analyze it in this article. As we shall discuss in more detail along
the case analyses, there are some differences among buying rms
regarding the SCF implementation. For instance, buying rms
may prioritize cash ows over automatization by focusing rather
on the extension of payment terms. Another rm might prioritize
the provision of exibility to its suppliers and provide them with
149
Figure 1: Supply chain nance (SCF) mechanism contrasting (A) transaction without SCF and (B) with SCF. Numbers are an illustrative example.
150
faceted perspectives that are necessary to understand the complexities of the innovation adoption process. In particular, we
address the following research questions:
By addressing these research questions, we make several contributions to research and practice. First, we provide in-depth
insights into the adoption of the innovative SCF process which
also provides further insights into upstream innovation diffusion
processes, highlighting necessary adjustments to the inuential
innovation adoption framework of Rogers (2003) to account for
the specicities of upstream innovations. Practitioners can gain
know-how on the SCF implementation process from an internal
and an upstream supply chain perspective reducing uncertainty
on how to implement SCF. Finally, we provide an empirical contribution to the emerging eld of research on the interface of
logistics and nance.
The remainder of this article is structured in six sections. First,
we present the literature review on SCF and (upstream) innovation adoption, concluding with our research framework. Subsequently, we describe our multiple case study method. Next, we
present the results of our cross-case analysis based on which we
developed testable propositions extending our initial research
framework. We conclude the article by discussing its theoretical
and practical implications, pointing to its limitations and suggesting paths for further research.
LITERATURE REVIEW AND CONCEPTUAL
FRAMEWORK
Two literature streams inform our research: (1) literature on the
logistics and nance interface and (2) relevant literature on upand downstream innovation management and organizational
innovation adoption.
SCF can be localized in the literature stream on the interface
between logistics and nance, which only recently gained
increased attention (Gupta and Dutta 2011). Most research has
thus far focused on conceptual work. For instance, Gomm
(2010, 135) denes SCF as a process optimi[z]ing the nancial
structure and the cash-ow within the supply chain, whereas
Grosse-Ruyken et al. (2011, 15) follow Camerinelli (2009) and
Pfohl and Gomm (2009) in dening SCF as an integrated
approach that provides visibility and control over all cash-related
processes within a supply chain. Our denition takes an
upstream supply chain perspective and focuses on the organizational structure to be implemented between the involved parties
to achieve visibility and control and to recurrently take cash ow
optimizing actions as outlined by the denitions presented
above.
Even though we tackle an empirical research phenomenon
there has been a rise in formal analytical modeling papers
informing our research on the logistics and nance interface
(e.g., Buzacott and Zhang 2004; Berling and Rosling 2005; Cal-
D. A. Wuttke et al.
dentey and Chen 2009). For instance, Protopappa-Sieke and Seifert (2010) analyze the interrelation of nancial and logistics
decisions within a supply chain indicating that improved
upstream cash ows may reduce suppliers nancial constraints
and thus improve physical ows. In addition, Shang et al. (2009)
study coordination schemes within supply chains and state that
SCF platforms provide the technology of applying such coordination mechanisms indirectly affecting logistics performance.
Although these models indicate that SCF has further benets
besides pure reductions of capital costs, they do not reveal
insights on how the SCF innovation can be established and disseminated by the buying rm. Hence, our article focuses on a
hitherto unexplored aspect of SCF.
The second stream informing our research on innovation adoption dates back to Ryan and Gross (1943) revealing that individuals face an innovation-adoption process consisting of several
stages of decision making (Ettlie 1980; Rogers 2003). According
to enhancements of these models, organizational innovation decision units must complete sequential process stages to nally
adopt an innovation (Meyer and Goes 1988; Rogers 2003). Analyzing innovation decisions along such frameworks has several
merits. In particular, by analyzing these stages separately, it
resolves the problem that some structural variables, such as centralization and organizational slack, may have a positive as well
as negative impact on the innovativeness of a rm (Rogers
2003). Furthermore, such models help to classify organizations
in more detail than by purely building upon the distinction of
adopters versus nonadopters (e.g., Meyer and Goes 1988; Fichman and Kemerer 1997). Therefore, in principle, such models
also inform the SCF adoption process. Particularly, the frequently
cited Rogers (2003) framework as depicted in Figure 2 appears
to serve as an appropriate foundation for structuring our research,
as it is generic enough to be transferred to SCF, but specic
enough to show how stages are interrelated, even though it
addresses predominantly organizational innovations opposed to
the upstream innovation in our case of SCF.
Moreover, the framework does not require the research team
to distinguish the degree of newness of SCF to the buyer and its
suppliers. The SCF concept might be initially new to the buying
rm and to early adopting suppliers, yet as the buying rm
reaches the routinizing stage (see Figure 2) and more suppliers
adopt SCF, the degree of newness to the buying rm decreases
until it becomes an established structure of buyersupplier relationships. Still, in the routinizing stage, the SCF innovation may
be new to late-adopting suppliers that board on to the already
established SCF platform of the buying rm (cf., Johannessen
et al. 2001).
Nevertheless, it is important to highlight the specicity of the
SCF innovation adoption by presenting to what extent ndings
from previous innovation research streams are or are not transferable to the SCF context. First, most of the innovation adoption
models were designed and tested in a within-rm context without
considering the upstream buyersupplier interfaces (Meyer and
Goes 1988; Ettlie and Reza 1992; Rogers 2003). Still, a specic
stream of research focused on the benets available from managing business processes across entities of the supply chain for faster product innovations, also termed design chain management
(e.g., Twigg 1995), or supplier involvement in new product
development (NPD) (e.g., Van Echtelt et al. 2008). Authors in
151
Figure 2: Innovation adoption stage model adapted from Rogers (2003, 421).
152
D. A. Wuttke et al.
Case selection
Data gathering
Data analysis
Construct
validity
Not applicable
Internal
validity
Not applicable
Multiple sources:
Questionnaires, semistructured interviews,
databases, and reports
Tandem interviews whenever
possible to reduce biases
Condentiality and
anonymity ensured
Multiple respondents
Most knowledgeable key
informants for SCF
interviewed
External
validity
Semi-structured interview
guidelines reported in case
study protocol
All interviews transcribed by
interviewers
Notes: References state where we adopted our procedure from. SCF, supply chain nance.
153
Coding
Table 2: Case demographics and stage relative to innovation-decision framework by Rogers (2003)
Cases
Alpha
Beta
Gamma
Delta
Epsilon
Zeta
Industry
Country
Size
S&Ps rating
First exposure to SCF
Highest management
level involved
State
Stage
Coatings
Netherlands
Large
BBB+
2010
Not applicable
Aviation
France
Large
BBB+
2009
Head of SC
Chemistry
Switzerland
Large
BBB!
2009
CPO
Pharmaceutical
Switzerland
Huge
AA!
2009
CEO
Chemistry
Germany
Huge
A!
2008
CEO
Automotive
Germany
Large
A!
2006
CEO
Initiation
Matching
(awareness)
Initiation
Matching
(potentially
rejecting)
1
1
1
Implementation
Restructuring/
redesigning
(evaluation)
1
2
3
Implementation
Restructuring/
redesigning
(piloting)
1
1
2
Implementation
Clarifying
(on-boarding
of suppliers)
1
3
3
Implementation
Routinizing
(continuing)
# interviews logistics
# interviews procurement
# interviews nance
1
1
1
1
2
2
154
D. A. Wuttke et al.
Codes
Original quotes
Redening
Benets allocation
Although the economic idea is the same in different SCF solutions we considered,
(to upstream
mechanism
they all differ in details. [] In the beginning, our priority was to nd a solution that
SC needs)
best ts our needs.
Degree of automation The system is very well integrated. [The supplier] uploads all required information to
the platform and we import these settings and simply ick a switch in our ERP
system., Suppliers have no longer issues with reconciliation as they always have
real time transparency whether we have released their invoices. Once we have done
so, they are informed automatically and can decide themselves when to get paid.
Scope of suppliers
We segment our suppliers and only target strategic or crucial ones. We do not intend
to onboard any leverage or opportunistic suppliers., As soon as the system was in
place, we tried to get as many suppliers on board as possible. However, those with
an annual spend below 200,000 are not interesting for our set-up. []
Restructuring
Cross-functional
We met weekly to discuss operational issues and twice a month with the whole SCF
collaboration
team., We needed to work very closely with our colleagues from nance and still
have formal weekly meetings., At the moment, we have a leadership team and we
discuss strategic SCF issues directly on top management level, It is crucial to
integrate operationalnot only strategicprocurement managers from the start on to
have better, immediate feedback from those who will work with the system.
Job design
Upstream supply chain managers initially said: Procurement managers initially said, it
is not our task to use SCF, it is not in our working contracts and job description. We
do not want to implement SCF
Performance
In our experience, SCF is likely to fail without changes of incentive structures., We
measurement
introduced working capital targets for direct and indirect materials. These were specied
at category level. There are different ways how logistics managers may achieve these
targets, but it is clear to date that working capital optimization and thus SCF adoption
along the supply base are among the most important ones.
Technology
It was quite tedious to adjust our ERP system to t to the SCF platform offered by the
[bank]. In total it took probably half a year., We developed a new payment platform
internally, anyway. When we learned about SCF, we discussed with our bank how we
could adjust our systems to become fully compatible. That was actually not too
complex., Moreover, we had to develop software internally and adjust our systems.
[] but in total, these efforts where not signicant.
Supplier
Suppliers opinions
We started recently two pilot projects in two European countries., We rst thought,
involvement
included in
SMEs would be primarily interested in participating [], but interestingly we were
decision making
wrongthat surprised us. In contrast, another suppliera large rmwas much more
eager to join the system, particularly due to off-balance sheet effects., We used the
experience gained in these pilot countries to improve our SCF solution to t to a
broader population of suppliers.
Feedback
The feedback from suppliers helped us to focus on such benets and to better
taken seriously
address suppliers needs.
Timeliness
As soon as we recognized the importance of suppliers in the process, we tried to include them
in pilot studies. The feedback of these selected rms was used to push the project forward.
Dissemination Suppliers
In most European countries, suppliers have not heard of SCF. So we had to explain
become aware
it to them., As soon as the rst manager of a supplier says sounds great, we are
interested, we have a rst success.
Suppliers become
We really need to go there and persuade suppliers that were not involved at an early
persuaded
stage of the project. We have workshops with their CFOs and explain the concept indepth to persuade them based on the benets both parties can obtain, Sometimes we
use our market power to persuade suppliers.
Suppliers adopt SCF Well, you know it is nally great to see that you are successful with the project and
suppliers adopt it. Recently, two or three really important suppliers in terms of spend
volume agreed to use SCF.
Continued.
155
Table 3: (Continued)
Category
Relationship
strength
Codes
Original quotes
Trust
Power
Communication
intensity
SCF leverage
Credit rating
Bank integration
Qualitative benets
Instead of us also the bank could approach our suppliers initially. However, suppliers dont
trust banks like they trust us. Due to long collaboration, these suppliers know us and once
we explain the idea, they say, okay, we are interested in the benets.
If we have more power and if we perceive that SCF is benecial for both parties, we expect
our suppliers to extend payment terms but offer them SCF in advance, to ensure they have
cheap access to short-term liquidity., When we are talking with smaller suppliers, where
we account for 3050% of their sales, we simply tell them to use SCF.
We work very closely with our colleagues from nance when we conduct workshops with
suppliers. Our nance colleagues are very important to assure our credibility in these
workshops. They are our experts regarding nancial and accounting issues and the
suppliers participating in such workshops usually bring their CFO to the table.,
With smaller suppliers, we usually use written communication or a procurement
manager negotiates with them one-on-one. [] We will extend payment terms anyway.
It is up to youtake our SCF offer or leave it.
For sure, our strong credit rating was required. Without it, we could have never offered
such good conditions to our suppliers., In all ratings we are graded very well, so we
bring a strong cost reduction argument to the table.
Our bank trusts us. We work together with them since more than 100 years and they know
our records and how we conduct business., Well, actually we are very grateful to our
bank, because due to its experience and our close collaboration, the bank was able to
assist us with many legal issues., The experience with our bank is extraordinary: The
collaboration was excellent, they are very experienced. They have very good procedures
and are very committed to the system as such and our collaboration in general.
One is quick with overlooking benets that cannot be measured in numbers. For instance,
the pure availability of a new source of funding might be more important for some
treasurers of our suppliers than whether they save some money., The SCF solution is
very well integrated in our system and thus offers many benets to us as well as the
suppliers. The payment process got even easier and our suppliers appreciate having more
exibility and information., One of our suppliers told us, that he estimates to save two
man-days, as the reconciliation is much easier for him. He has quite some benets even if
he does not use the credit, because he always knows immediately when his
invoice has been released.
Implementation of SCF
The effectiveness of innovation adoption depends on the t
between the organization and the innovation itself (Van de Ven
1986). This t can be achieved in three ways: by restructuring
the organization, by redesigning the innovation, or a mixture of
both (Rogers 2003). We will now describe our observations with
respect to redening and restructuring, analyze common patterns
among all cases, and propose three generalizations. For summaries of codes and original quotes, please refer to Table 3.
Redening
Redening SCF is a process during which rms not only adjust
the SCF innovation to their specic context but also initiate a
process in which rms need to reconsider the contextual factors
to which the SCF innovation needs to t (Johns 2006). The valuation of these factors varies over time, as we learn. Epsilons
nancial manager claimed: Although the economic idea is the
same in different SCF solutions we considered, they all differ in
details. [] In the beginning, our priority was to nd a solution
156
that best ts our needs. We learn that the value added for suppliers gains increasing importance during the adoption process.
Epsilons procurement manager argues as follows: If SCF serves
only Epsilons needs, but not the supplier needs, why should any
supplier agree to use SCF? So, Epsilon analyzed how SCF can
be congured to t supplier needs. One aspect of this adjustment
is the particular focus on the provision of visibility into cash
ows and invoice releases such that the suppliers of Epsilon
obtain the maximum degree of exibility, as they can better plan
when they will receive cash ows. Similarly, Delta and Zeta
found solutions to particularly serve supplier needs. Another
aspect of the redesigning of the innovation is revealed by Delta
as it builds its SCF implementation on an already existing automated payment platform within its corporate group. From a technical and information system perspective, Delta thus only needed
to adjust this platform, rather than to integrate an entirely new
one. In this way, Delta redesigned the SCF innovation to its own
needs, rather than having implemented an external standard solution. Thus, rms go through a process of dening criteria to
which the SCF innovation needs to t, before they can actually
redene SCF.
A pattern that emerged among all cases is that there are three
broad categories of redening SCF which rms need to consider
to make context tting decisions: (1) benets allocation mechanisms, (2) degree of automation of transactions, and (3) scope of
suppliers using SCF, summarized in Figure 3.
Restructuring
The SCF innovation requires cross-functional collaboration of
nance and logistics as well as procurement in a way which was
new to all our sample rms. This becomes evident from Epsilons nancial manager who stated: As I never had any operational contact to suppliers I rst had to get to know our internal
managers for suppliers. To not know the own colleagues from
the upstream supply chain functions indicates the absolute
absence of any form of direct communication, making direct collaboration infeasible (Grant 1996; Kahn and McDonough 1997).
But, active collaboration between nance as well as logistics and
procurement managers is required to adopt SCF successfully
because the former are dedicated nancial experts and the latter
often manage the interface with suppliers (Stefansson and Russell
2008). So, the rst step was to bring together responsible
managers of each function in the case of Epsilon. From initial
occasional project meetings, quite formalized patterns of collaboration emerged. At Epsilon, we observed weekly formalized
meetings to discuss recent issues, but also informal meetings
whenever necessary indicating cross-functional collaboration
(Pinto et al. 1993; Kahn and McDonough 1997).
Moreover, adjustments with respect to job design had to be
made. Particularly in rm Zeta, conicts arose due to new tasks
inherent to SCF, as its nance manager explained: Procurement
managers initially said, it is not our task to use SCF, it is not in
our working contracts and job description. We do not want to
implement SCF. In contrast to traditional tasks of upstream
logistics and procurement managers (Chopra and Meindl 2012),
the adoption of SCF requires further tasks as it becomes necessary to market the SCF innovation upstream. These marketing
skills, however, are quite distinct from traditional skills present
in these functions (Ellinger et al. 2000). Typically, in the case of
D. A. Wuttke et al.
157
ment. With this term, we indicate that up-to all three functions
need to be aligned toward common goals of SCF while we
acknowledge that whether or not logistics and/or procurement
needs to be aligned with the nance function depends on the
organizational structure, as Epsilon and Zeta show. Such an
alignment differs from the prior working relationship between
the three functions in all analyzed rms. Whereas before adopting SCF members of these three functions essentially only agreed
that costs have to be reduced (only procurement and logistics
have certainly more congruence), logistics/procurement-nance
alignment implies that all of these functions share the same
objectives and think in the same direction toward SCF aims
based on a broader mindset. The involved functions need to
agree that a practice such as SCF can actually improve corporate
performance, although it is distinct from the traditional tasks of
logistics and procurement. On the other hand, nancial managers
involved in the SCF project need to understand challenges in the
current supply chain setting, before deciding on how to use SCF.
Yet, logistics/procurement-nance alignment does not per se lead
to SCF adoption. Consider for example Delta, which recently
gained quite high-internal alignment regarding SCF, but is still in
early stages of the implementation process. Nevertheless, logistics/procurement-nance alignment strengthens the link between
redening and restructuring, and thus indirectly has a positive
impact on the effectiveness of SCF implementation.
Second, a similar pattern emerged with respect to early
involvement of suppliers. Delta, Epsilon, and Zeta have only
concentrated on the internal perspective in the very beginning,
but soon recognized the need to include the interest of suppliers
as well. Therefore, they selected suppliers for pilot projects, conducted intensive workshops, and remained open for serious feedback of suppliers. The objective of such an involvement is to get
meaningful feedback on SCF for a dened time during the adoption project to redene SCF according to the suppliers needs.
This is opposed to the previously studied supplier integration for
recurring interaction focusing on the joint set-up of infrastructure
for efciency in recurring NPD projects with preferred suppliers
(Monczka and Morgan 1996; Das et al. 2006). A consequence
resulting from this difference is manifested in the selection of
suppliers for early involvement in the restructuring and redening stages. In the case of Delta, we learned that important, but
not crucial suppliers were selected for these studies, as Delta was
afraid of unpleasant side-effects such as endless workshops
distracting a suppliers valuable resources from required business. Furthermore, these suppliers were purposefully selected:
158
Proposition 3
Clarifying and disseminating are interrelated and mutually
enforcing processes in the sense that neither alone can be
successful without the other process advancing.
This stage of dissemination is distinct from previous research as
the role of external members in the innovation process is much
smaller in common organizational innovations (Van de Ven 1986;
Rogers 2003). Therefore, our analysis required us to go beyond
our knowledge about previous innovations and to explore attributes increasing the effectiveness of rms during the dissemination
stage. In contrast to previous downstream innovation dissemination, the focus of SCF is more on persuading the right people than
on maintaining and improving interorganizational relationships by
specic long-term investments to develop new products (Kim
2000). Note, that both tasks are quite distinct.
Effectiveness of the upstream dissemination process
Suppliers are willing to adopt SCF, if they expect high returns,
which is central for the assessment of a buying rms dissemination effectiveness. Although a buying rms credit rating determines in principle the interest rate a supplier would face using
D. A. Wuttke et al.
where nancial experts from both parties are involved. In particular, we see the use of the latter rather with suppliers of high
importance.
These three attributestrust, buyer-power, and communication
obtrusivenessreect categories of relationship strength, as each
of them is relationship specic opposed to SCF leverage. The
more buyersupplier relationships are characterized by such
strengths, the more effectively a buying rm can disseminate
SCF in the supply base. More formally,
Proposition 4b
Relational strength has a positive impact on the dissemination of SCF in the sense that it increases the effectiveness
of the dissemination process in the supply base.
Propositions 4a and 4b state that buying rms have particularly four concepts for persuading suppliers, namely SCF leverage, trust, power, and communication obtrusiveness. These
concepts determine the enforceability of dissemination (i.e., the
persuasiveness of the buying rm). But they differ clearly in
their results: although a concept such as buyer power virtually
forces weaker suppliers to adopt SCF, the other extreme of
purely offering benets might be less effective as discussed
above. Another important dimension is the long-term buyersupplier relationship quality. In contrast to the enforceability of dissemination, offering true benets to the supplier has a positive
impact on the buyersupplier relationship quality (Ganesan
1994), while the use of coercive power might deteriorate supplier
satisfaction (Benton and Maloni 2005). Between both extreme
poles are trust and communication obtrusiveness. Although both
are slightly more enforcing than purely suggesting benets, their
overuse may deteriorate the relationship quality as well. As communication becomes too penetrative, suppliers may generate a
feeling of resentment.
This trade-off between enforceability of dissemination and the
quality of buyersupplier relationships uncovers a specic
dilemma of upstream innovation dissemination. Although SCF
usually benets the supplier, there might be resistance or inertia in
the supplier base, which might induce buying rms to employ
more effective concepts of persuasion deliberately risking a
decrease in buyersupplier relationship quality. This is different in
downstream dissemination where often customers actively demand
innovations (Neale and Corkindale 1998; Gruner and Homburg
2000). Therefore, in the downstream dissemination achieving high
enforceability and a positive impact on relationships is often not a
contradiction opposed to upstream dissemination. The effect of
concepts used for accelerating the upstream dissemination process
on the enforceability of dissemination and the buyersupplier relationship quality are characterized by Figure 4.
159
160
D. A. Wuttke et al.
Figure 4: Enforceability and relationship impact of dissemination concepts of supply chain nance (SCF).
161
162
COMPANY ALPHA
Alpha is a multinational coating and chemical company.
Although it supplies both, industry and consumers, we focus on
its industrial coatings sector. Alpha has a very strong reputation
for sustainable business practices requiring innovation. The
nancial crisis heavily affected the specialty chemical as well as
the coatings industry. Supplier defaults as a consequence of
liquidity shortages were, however, rather the exception.
Despite being innovative in products, Alpha is still in a very
early stage of the SCF adoption. Alphas managers were only
becoming aware of SCF and its benet. Alpha still hesitates to
implement SCF as many suppliers are from Northern Europe,
particularly the United Kingdom where payment terms are relatively short. Therefore, these managers believe that potential benets of SCF might be rather small. On the other hand, Alphas
managers are concerned about bringing the innovation through
the supply chain, as other matters are of higher importance for
its suppliers than nancing like sustainability projects. Therefore,
Alphas managers are afraid that enforcing SCF too heavily in
the supply base might have negative consequences.
Thus, Alpha can be considered being in the matching stage as
it tries to identify how SCF could help to meet its logistics goals
of further inventory optimization along its supply chain.
COMPANY BETA
Beta is a globally leading aircraft manufacturer. The aviation
industry is characterized by high levels of safety standards making
it almost impossible to exchange strategic suppliers. A further
characteristic of this industry is that many products are prenanced
by the buyer including advance payments. Thus demanding payment terms extensions for strategic suppliers is rather unusual for
Beta. At the moment, Beta is trying to improve its working capital
through consignment stock and vendor managed inventory (VMI),
as, during the nancial crisis working capital gained increasing
importance. Therefore, the reduction of working capital, also with
a strong cash focus, is on the top of Betas corporate agenda.
Betas managers, at the time of data collection, have been
gathering information from diverse banks to understand how
SCF could t to its particular supply chain and industry setting.
However, against the background of high safety standards, Betas
managers appear also quite riskaverse when it comes to implementing process innovations such as SCF. Therefore, Beta managers believed that it was better to wait for other rms adopting
SCF to avoid early adopter risks. Moreover, Beta believes that it
will depend more and more on nancially healthy suppliers in
the future. Thus, its suppliers should rather be able to nance
themselves than relying on solutions like SCF. Finally, there is a
lack of top-management commitment regarding SCF, as other
working-capital related projects such as VMI have been recently
launched.
Hence, Beta can be considered in the matching phase, but is
potentially rejecting SCF.
D. A. Wuttke et al.
COMPANY GAMMA
Gamma is in the chemical industry supplying mainly industrial
rms. During the nancial crisis it suffered heavily from industry
downturn. Although supplier defaults were exceptions, Gammas
managers perceive that many of its suppliers have nancial constraints and might be negatively affected by extended payment
terms of Gamma.
Gamma has recently launched several initiatives targeting the
improvement of working capital. Thereby, it approaches the cash
conversion cycle holistically: After signicant reduction of inventories, it strived for the systematic extension of payment terms
with its suppliers, and reduction of payment terms with its customers. Whereas these measures allowed substantial working
capital improvement, management is now concerned that further
improvements are difcult. Therefore, top managers of Gamma
are continuously urged to identify new means of improving net
working capital.
In 2009, Gamma rst got exposed to SCF, as it learned from
successful business cases in other industries, mainly the automotive industry. Due to its centralized cash management team, SCF
obtained quite fast top management commitment. Within the corporate team, Gamma identied soon how SCF could potentially
be matched with its own needs and how it could be employed.
However, the team also faced immense challenges implementing
it. For instance, nancial auditors mentioned that SCF might
result in classication conicts in auditing, as under certain circumstances amounts handled through SCF could no longer be
classied as accounts payables, thus destroying the working capital effects of SCF. Furthermore, Gamma has a relatively weak
credit rating in comparison to other rms in our sample, which
might be still better than some of its suppliers, but managers perceive a lower benet due to the credit rating.
As Gamma found no satisfying solution it still tries to restructure SCF according to its own needs.
COMPANY DELTA
Delta is an innovative company in the pharmaceutical industry.
In this industry, margins for patented products are quite high
even during the nancial crisis. Particularities of the industry are
the extremely high requirements on product availability because
production stops might threaten lives. Therefore, Deltas highest
supply chain objective is a stable supply of raw materials. Working capital optimization and payment terms are rather secondary
targets. However, during the nancial crisis certain suppliers
faced immense nancial shortages. As one manager of Delta
states, the question was no longer whether suppliers were able
to deliver the right quality at the right time, but rather whether
they would be nancially able to deliver at all. Therefore,
increasing attention was placed on working capital management,
particularly with respect to suppliers.
Independent from its supply chain, Delta started a payment centralization project in close collaboration with its corporate bank in
2009. The aim of this project was to increase transparency of its
payment processes and to pool cash resources on a company-wide
scale. This project required substantial changes in its internal platforms, as Delta is a huge, global rm with dispersed subsidiaries.
163
extended payment terms, but offering SCF as a means of reducing the resulting burden to the supplier.
Epsilon is content with the already achieved results of the
SCF solution. Therefore, the plan of Epsilon is to bring SCF into
a routinizing state now, to use it in its daily processes, where
on-boarding new suppliers will rather be an occasional
exception.
COMPANY ZETA
Zeta is a German manufacturing company in the automotive
industry facing a highly competitive environment with enormous
pressure on costs and working capital. The pressure on costs
even increased during the nancial crisis, when customer demand
was signicantly smaller and capacities could not be fully utilized. Zeta is known for innovative and high quality products.
Zeta, although it produces in Europe where it faces rather
short payment terms with its suppliers, delivers to Asian rms
which usually have long payment terms. Therefore, on average it
pays its suppliers 20 days earlier than it is paid by its customers.
However, Zeta achieved a level of payment term extensions with
its suppliers where it would be difcult to further extent the
terms without facing serious liquidity problems of suppliers. So
even before the nancial crisis, already in 2006, working capital
management received increasing attention. At that time, Zeta
identied the practice called SCF to be successful outside of Europe, but yet did not nd any solution provider in Germany.
Therefore, it started collaborating with its corporate bank, which
so far had no experience with SCF in Germany, to set up a new
SCF platform. This project was initiated in early 2007 and
shaped by mutual learning between Zeta and its bank.
Early implementation stages were characterized by numerous
internal discussions and several struggles caused by a low willingness to adopt a new process. However, when top management
commitment raised, communication improved and rst positive
cases highlighted the acceptance grew. Since the two-rst pilots
in the end of 2007, an increasing number of suppliers switched
to SCF. After a successful adoption of SCF in Germany, further
European countries were targeted and suppliers on-boarded. By
the end of 2011 almost all important suppliers, which were
intended to be switched, have been addressed and invited to use
SCF. Many of them agreed.
So, eventually SCF is becoming a routine for the procurement
managers. When asked about SCF in 2011, one procurement
manager answered us: SCF is a standard tool, which we use to
avoid negative consequences of extended payment terms for our
suppliers. This clearly highlights that after Zetas long way of
redesigning its processes and redening the characteristics of the
SCF platform, today SCF is a routine process.
REFERENCES
Aberdeen Group. 2006. New Paradigm Supply Chain Finance
Offerings Compel CFO and Treasury Interest in the Supply
Chain. Boston: Aberdeen Group.
Aberdeen Group. 2007. The 2008 State of the Market in Supply
Chain Finance. Boston: Aberdeen Group.
164
D. A. Wuttke et al.
165
166
D. A. Wuttke et al.
has taught strategic sourcing and procurement at several international universities (e.g., IAE Aix-en-Provence, Lomonossov University Moscow, St. Petersburg State University, Technical
University Berlin) and in numerous Executive Education programs (e.g., BASF, Siemens, DHL, Nestl!e, Deutsche Bank). At
IIM Bangalore he helped to build up the rst chair for strategic
sourcing and procurement and hosted the rst India Sourcing
Summit with international exposure. In addition, Dr. Blome is
actively involved in the international research scene as a frequent
presenter at Academy of Management, POMS, EUROMA, Decision Sciences and IPSERA conferences. His research interests
are all related to the strategic sourcing and procurement domain
where he tries to always explore the most recent phenomena to
provide future guidance to the discipline. Up to now, his
research has been published in international journals like Journal
of Supply Chain Management, International Journal of Production Economics, International Journal of Production Research,
Journal of Business Logistics, Journal of Business Ethics, and
Journal of Purchasing & Supply Management. Currently, Dr.
Blome supervises around 10 PhD students.
Constantin Blome (Dr. rer. Oec. Technical University of Berlin) was appointed as the GSK Vaccines Chair Professor in Strategic Sourcing and Procurement at Universit!e catholique de
Louvain (UCL) in September 2011. Dr. Blome has gained substantial experience in sourcing and procurement in various ways.
To name a few: He was senior consultant in the procurement
domain for a Swiss consulting rm. Following his doctoral
degree in 2006, was hired as an Assistant Professor in Sourcing
in Emerging Markets at the EBS Business School, Wiesbaden,
Germany where he contributed in growing the Supply Chain
Management Institute to one of the largest and most successful
European research groups in this eld. Furthermore, Dr. Blome
Michael Henke (PhD Technical University in Munich/Germany) nished his habilitation in 2007. Prior to joining EBS as a
Senior Professor, he worked as Senior Consultant with the Supply Management Group in St. Gallen, Switzerland. In his current
role Michael Henke is responsible for teaching and research in
the areas of nancial and management accounting, supply management, supply risk management, supply performance management, supply controlling, and others. Additionally, he is leading
and developing various research projects in the eld of nancial
supply chain management in cooperation with well-known international enterprises. His research ndings can be found in
numerous journal articles, book sections as well as books.
SHORT BIOGRAPHIES
he ndings from an inductive study conducted in the United States and Europe focused on sustainability implementation efforts across
supply chains are reported. In particular, the study focuses on developing a better understanding of: how do companies involve other
members of their supply chain into their sustainability efforts? Building upon themes that emerged from the data, a typology for the supply
chain implementation of sustainability initiatives is proposed. A lack of supply chain integration was identied and companies reported a tendency toward a mandated implementation when extending efforts across companies. These efforts are generally initiated by the dominant
companies and then forced onto the weaker upstream members. The data suggests that the merits of these initiatives are viewed as being disproportionately awarded to the dominant rm and thus did not receive full buy-in from the other party. A common vision by both sides of this
relationship is to develop a more collaborative implementation that can be supported by all the involved parties. In theoretical implications,
empirical ndings are viewed through the lens of several prominent management theories in order to augment and elaborate current theory.
Managerial implications, limitations, and opportunities for further research are detailed.
Keywords: sustainability; supply chain integration; collaboration
Corresponding author:
Sebastian Brockhaus, Institute of Business Logistics and General Management, Hamburg University of Technology, Schwarzenbergstrae 95,
21073 Hamburg, Germany; E-mail: sebastian.brockhaus@tuhh.de
168
the four walls of an individual company and the formation of relationships in a supply chain, only a limited amount of research has
focused on the interplay between companies as it relates specically to sustainability efforts.
In the next section, the extant literature is presented and then
offered as motivation for addressing the issue of how to implement sustainability efforts from a supply chain perspective. Following this, the methodological process used in this research is
described. Next, the ndings of the analysis are presented,
accompanied by relevant excerpts from the qualitative data,
which reect emerging concepts. This section establishes a typology for implementing sustainability initiatives across ones supply chain and makes propositions for further inquiry. This is
followed by a presentation of the theoretical and managerial
implications of the ndings from the study, along with limitations and future research opportunities. Finally, conclusions are
drawn.
LITERATURE REVIEW
While the concept of sustainability has been around for decades,
a clearer understanding of the topic has only emerged in the past
20 years (Haugh and Talwar 2010). As sustainability has been
discussed in several publications in recent years, the point of this
research was to focus on the implementation of the concept
across the network of companies that represent a supply chain
and expand existing theory surrounding this specic issue as theoretical gaps can be identied in this context (Carter and Rogers
2008).
Green supply chain management
There are several streams of literature that cover the role of sustainability in supply chain management (SCM). One of the major
areas is the eld generally termed green SCM (GSCM). As the
term suggests, the focus here is on the environmental aspects of
supply chains. The literature reviews by Srivastava (2007) and
Abukhader and Jonson (2004) provide a good overview of this
stream of research. Both articles point out that GSCM is mainly
derived from a reverse logistics angle, but also has inputs from
concepts focusing on overall process efciency. Reverse logistics
describes the upstream ow of resources in combination with a
reduction in materials to make the transport more efcient (Carter and Ellram 1998). Reverse logistics activities play an important role with respect to the environmental impact of supply
chains and have received amplied attention in the literature. For
a current review of work in this area, see Chan et al. (2010).
Srivastava (2007) examines the synergy of the ecological and
economic aspects of SCM. This notion has been promoted by
several authors before when describing the role of green for
competitive advantage (e.g., Porter and Van der Linde 1995a,b;
Florida 1996; Rao and Holt 2005; Kersten et al. 2010; Mollenkopf et al. 2010). This literature suggests that resource efciency leads to a reduction in operating costs as well as having a
positive environmental impact and thus strengthens the competitive position of the enterprise. However, as Mollenkopf et al.
(2010) and Kersten et al. (2010) among others caution, creating
genuine and long-lasting competitive advantage for the entire
S. Brockhaus et al.
169
14000 is ill-equipped to foster diffusion of the sustainability concept across companies. Potoski and Prakash (2005) demonstrate
that ISO 14000 lends itself well to a reduction in the necessary
effort for compliance with environmental and social regulation,
but is not of much benet to implementing sustainability outside
of compliance issues, thus falling short of holistic SSCM implementation.
Connelly et al. (2011b) argue that the theoretical foundations
for sustainability in businesses are comprised of several theories.
Similarly, Carter and Easton (2011) explicitly recommend the
use of multiple theories when engaging in research around sustainability to improve the depth of the analysis. In a recent piece,
Seuring (2011) argues that SSCM theory changes the application
of general SCM theory as certain criteria and assumptions (e.g.,
concerning sourcing decisions) are altered by expanding the view
from the single to the TBL. Thus, current management theories
will augment this studys data in support of deriving a typology
for SSCM implementation. A selection of theories suggested by
Carter and Rogers (2008), Connelly et al. (2011b), and Carter
and Easton (2011) with high relevance for SSCM is presented in
Table 1. The selection of theoretical lenses displayed in Table 1
will be applied in the description of this studys ndings to
derive theoretically sound implications. The ndings are mirrored
with theory to gain further insights into the relevance of the ndings as well as to provide suggested advancements of the theory
based on the analysis.
METHODOLOGY
As established through the literature review, the theory around
implementing SSCM is still abstract and broad in nature (Faber
et al. 2005). As argued by Lubin and Esty (2010), this results in
implementation of sustainability efforts that are characterized by
a lack of structure. Thus, consistent with the suggestions of
Fawcett and Waller (2011), this study was conducted to seek
clarity around sustainability implementation in companies and
supply chains rather than conrm preconceived ideas. Starting
from the empirical data, a theoretical framework is derived and
then mirrored with existing theories and literature. A qualitative
research design (Easterby-Smith et al. 2002; Frankel et al. 2005)
was deemed appropriate given the goals of the study. In particular, a grounded theory (GT) approach is taken, based on the fact
that it is designed to explore complex phenomena in real-life situations (Glaser and Strauss 1967). As the implementation of the
theoretical concept of sustainability across supply chains is being
investigated and the authors have a strong desire to conduct
research of high managerial relevance, GT suits the topic in
question (Glaser 1999; Mello and Flint 2009).
GT was established by Glaser and Strauss (1967) and further
rened by many authors, for example Glaser (1978), Goulding
(1999, 2000, 2002), Corbin and Strauss (1990, 2008) and Charmaz
(2006). Following the suggestions of Pratt (2008, 2009), an explanation of how sampling, data collection, and analysis were conducted and how research validity as well as ties to existing theory
was ensured is provided. Further guidance on how to employ GT
for business logistics and economics research beyond the scope of
this section can be found in Mello and Flint (2009), Maital et al.
(2008), Flint et al. (2005), Wilson (2002) and Finch (2002).
170
S. Brockhaus et al.
Stakeholder
theory
Resource
dependence
theory
Transaction
cost
economics
Signaling
theory
Sources
171
Industry
Region
Respondent position/department
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Consulting
Food and Retail
Consumer Electronics
Retail
Chemicals
Food
Food
Food and Retail
Logistics
Logistics
Logistics
Retail
Logistics
Industrial Products
Paper
Apparel
Home Appliances
Medical Supplies
Food
Food
Automotive
Logistics
Food
Consulting
Food
Industrial Products
Food
Sporting Goods
Europe
Europe
Europe
Europe
Europe
Europe
Europe
U.S.
Europe
Europe
Europe
Europe
U.S.
U.S.
Europe
U.S.
U.S.
U.S.
U.S.
U.S.
U.S.
U.S.
U.S.
U.S.
U.S.
U.S.
U.S.
U.S.
Consultant
(1) Head of Sustainability and (2) Representative from the Logistics Department
General Manager Logistics
Head of Logistics
Supply Chain Management
Chairman of Executive Board
(1 and 2) Logistics Management, and (3) logistics service providers of interview partner
(1) Customer Service, (2) Head of Logistics, and (3) Head of Sustainability
Global Head of Green Logistics
(1) Senior Director Logistics and (2) Head Business Development
Head of Green Logistics
Division Management CSR
Head of Sustainability
Vice President of Global Sustainability
Head of Group R&D
Head of Logistics and Supply Chain Management (SCM)
(1) Head of Sustainability and (2) Logistics and Supply Chain
(1) Category Management and (2) Warehouse and Logistics
Supply Chain Management
Product Design
Product Design
(1) Human Resources and (2) Internal Consultant
Head of Logistics and SCM
Consultant
(1) Supply Chain Management and (2) Product Design and Packaging
Logistics and Supply Chain
Corporate Social Responsibility
Operations and Logistics
172
S. Brockhaus et al.
Documentation
Detailed documentation of data analysis and rigorous reevaluation of codes with multiple
researchers.
Use of software to structure and document the analysis.
Concepts that emerged throughout the interviews were discussed with following participants.
Triangulation
Credibility
Results must be adequate
representations of the data
Transferability
Applicability of results to
larger population
Theoretical sampling
Only reoccurring themes were transferred into theory
Dependability
Stability and consistency
of results regardless of time and place
The collective experience of all participants was tapped into for the study, providing
a large body of know-how
International nature of the study reduces strong regional
impact
International research team reduces potential regional researcher bias
Long data collection process ensures avoidance of ad-hoc conclusions
Conrmability
Results are grounded in
actual data versus driven by
researcher bias
Researcher triangulation
Feedback of results with participants
Integrity
Results are inuenced by
reality and not intentionally misguided
by participants
173
Table 4: (Continued)
Criteria
Fit
The results are signicant and
meaningful for the topic under
investigation
Understanding
The respondents understand the
conclusions derived by research team
A summary of the interviews was given to all participants as well as colleagues and
practitioners for review
Working results were presented on conferences (POMS
2010) and workshops (workshop about sustainable products at Indiana University,
14th15th October 2010) and feedback was included
Generality
The ndings cover the subject
under investigation in a holistic
fashion and include multiple
dimensions of the topic
Semistructured interviews were open and of sufcient depth to allow for multiple
aspects of the issue to be discussed
Interview guideline was adapted to include further aspects as pointed out by the
participants
Control
Participants have an inuence
on developed theory
174
S. Brockhaus et al.
hey you know for us it is a very important topic and by the way,
reducing your energy input could even save your money (). So
it is mostly a matter of us persuading and convincing them to
tackle some of the points that we show them (Interview # 12,
Retail, Europe). Organized in the form of dyads, this cooperation
only involves companies in subsequent tiers in the supply chain.
It was the perception of the participants in the study that the
initiatives were implemented in this way across the supply chain
because only the stronger and more powerful members of the
chain could initiate such initiatives as this participant explains:
our ability to push an agenda on sustainability to our retailers is
relatively small. That is the same environment that exists with our
suppliers. So it is a pull process. And it starts with leadership
and the folks who have the leverage to get things done. Clearly
speaking, most of our suppliers do not have that leverage (Interview # 14, Industrial Products, U.S.). Based on the current statements (see also Table 5 at the end of this section) and the
aggregated data analysis, the following proposition emerges:
Proposition 1
Sustainability efforts are currently implemented in a mandated fashion as a pull process through the supply chain
and initiated by the stronger members of the chain.
Mandated sustainability initiatives are usually implemented on
a formal here is what we need you to do basis. A good
example of this is the use of codes of conduct a retailer requires
their manufacturer, supplier, or LSP to sign to establish a business relationship as this participant notes: usually the customers
have a policy we have to sign or we ll out a questionnaire.
There is not really a discussion (Interview # 9, Logistics, Europe). This form of implementation can be labeled mandated
because it does not involve much two-way conversation or
sharing of best practice or experience as the following participant
175
I would say, at this point it is on our end. We have a lot of reaching out to suppliers to do () these are all
things that we have initiated. (Interview # 18, Medical Supplies, U.S.)
From a supplier perspective, they have never come to us. It is us coming to them. Weve had no push from
the supplier perspective. So its this trickle-down effect, we are a big supplier to some retailers, who have put
us in their high-risk, highly monitored kind of group and we have replicated that process with our supply
base. But generally speaking, our suppliers are not pushing this topic at all. (Interview # 14, Industrial
Products, U.S.)
The bigger companies can pretty much get suppliers to create whatever they want or at least consider it.
With us, if it is not something that is emerging or readily available they are not creating it. Only in very
few cases, folks are going to create something special for us because of our [small] size. (Interview # 23,
Food, U.S.)
Further support
for Proposition 2
Our suppliers are always involved and quite frankly, we are very tough on our suppliers. We do supplier
challenges. You go out to current suppliers and potentially new suppliers and tell them what you are
looking for. () We denitely take advantage of the resourcing, insights and capabilities that our supplier
have. (Interview # 20, Food, U.S.)
We havent had a lot of success to be honest with you. The suppliers will call and promise the world and
when you sit down and dig down, in every case weve had so far, they are not really much further ahead
than the folks we are dealing with now. There is a lot of greenwashing going on and a lot of
self-promoting. (Interview # 23, Food, U.S.).
Further support
for Proposition 3
I was invited to a sustainability workshop at a company recently and they just kept talking about prices
for the logistics lanes in their last tender. Then somebody got up and asked if this was about sustainability
or about prices and they admitted that is was just pricing negotiations but they knew wed show up for
sustainability talks (Interview # 7, logistics service providers of interview partner, Europe).
Something that deeply impressed me, () there were customers who had been evaluating the company
and conducted a benchmark and we were in the top segment environmentallyand then nally they give
you no business and make their contracts with the cheapest company. (Interview # 10, Logistics, Europe)
Further support
for Proposition 4
We are moving to this notion of shared benets and shared responsibility and that is really the key to a
good relationship with any outside vendor. () As far as hiding the benets of those things in
partnerships, that doesnt work. So we are really trying to roll that out across the company, this notion of
shared benets. (Interview # 8, Food and Retail, U.S.)
In terms of collaboration, what I am seeing this far is that a lot of companies want to get together and
not just discuss: what is our sustainability strategy? But they want to look at what organizations are
doing, independently or collectively to be more sustainable and seek opportunities to collaborate.
I think one other plus side from just leaning your operation () sustainability in a macro view really is
about collaboration on one level in order to agree upon the processes, procedures and technologies that
are used to move products around the globe. (Interview # 13, Logistics, U.S.)
I think what I would do () is horizontal integration. Because I think for example trucks in Europe are
on average only 2030% full of cargo. So a huge opportunity lies in collaborating in the supply chain.
But obviously there are severe obstacles; companies do not want to share data or any information about
their supply chain. So I think there is a lot of work to be done in this area. (Interview # 9, Logistics, Europe)
if I sat at my desk alone, I would not have these ideas, but if you go out to talk with carriers and service
providers, industry at conferences or whatever, talk to your own people; the network in Europe is big enough;
then you get new ideas and some of them are really great. (Interview # 3, Consumer Electronics, Europe)
176
Proposition 2
Mandated sustainability implementations are characterized
by a lack of communication and collaborative behavior.
S. Brockhaus et al.
their warehouses: the retailer we worked with wanted the cost reductions passed on, that is the underlying thing (Interview # 27, Food,
U.S.). These ndings lead to the following research proposition:
Proposition 3
177
178
S. Brockhaus et al.
(2010) and Flint et al. (2005), the propositions that emerged from
the ndings serve as an impetus for follow-up research.
Opportunities also emerge for researchers to empirically analyze sustainability in the supply chain across different industries.
Given the wide focus of the article, the analysis did not focus on
looking at potential industry differences. Analysis in this direction may prove to be valuable as differences between industry
sectors and market types can be expected.
When companies extend their sustainability efforts into their
supply chain, the data indicate a mandated implementation
approach to be prevalent. There were strong indications that this
approach may lead to suboptimal results due to misaligned
incentives. The research proposition can be employed as starting
points for hypotheses examining this idea. Propositions 1 and 2
can be used for a quantitative validation of the mandated implementation of sustainability initiatives. Propositions 3 and 4 provide possibilities to analyze the current misalignment of the
incentives in the supply chain as well as develop approaches for
collaborative initiatives.
Based on the collected data and in accordance with TCE and ST,
collaborative work in sustainability is recommended as described
previously and companies are encouraged to press for a fair distribution of benets, creating winwin situations for themselves and
other members of their supply chain. This study provides initial
ideas for possible relationships between collaboration and transaction costs that should be further pursued empirically as well as
using modeling approaches. Researchers may also want to consider
engaging in research around the fundamental question, in what
ways intensied collaboration with suppliers as well as customers
can enhance the TBL performance of a focal company.
In accordance with general relationship models (e.g., Lambert
et al. 2009), companies should carefully evaluate and select their
strategically important business partners and approach them with
collaborative sustainability opportunities. In terms of potential
future research, the need for additional insights into the supply
chain dynamics with regard to sustainability emerges. Researchers can provide comprehensive models on how to structure relationships in the supply chain to address the problems of
incentive alignment. Partnership models should be expanded to
include sustainability issues and provide direction for appropriate
ways to allocate benets and risks across the supply chain.
CONCLUSIONS
Based on the ndings and the theoretical and managerial implications, it can be concluded that SSCM implementation is at an early
stage of development. Companies have yet to embrace the idea of
fully implementing sustainability into their supply chain relationships. The current dominant perspective to focus internally with
respect to sustainability cannot produce satisfactory results in terms
of TBL performance in the long term. Furthermore, when companies do extend sustainability efforts into the supply chain, they typically take a mandated implementation, which was viewed by
several companies as a suboptimal approach. Both from a theoretical point of view and a view supported by the current data, a more
collaborative supply chain implementation approach to sustainability, featuring long-term relationships, supplier development, and
fair distributions of incurring burdens and benets promises
enhanced results on all three bottom lines. This conclusion is further supported by the fact that the majority of the respondents indicated that sustainability efforts should not be considered as only a
fad or a passing phase, but as a major business trend that will continue to evolve in the years to come. Therefore, companies are
encouraged to actively seek out sustainability opportunities and be
willing to take risks associated with these efforts. Just like any business trend sustainability requires know-how that can only be
acquired through experience.
REFERENCES
Abukhader, S.M., and J
onson, G. 2004. Logistics and the
Environment: Is it an Established Subject? International
Journal of Logistics: Research & Applications 7(2):13749.
Agle, B.R., Donaldson, T., Freeman, R.E., Jensen, M.C., Mitchell,
R.K., and Wood, D.J. 2008. Dialogue: Toward Superior
Stakeholder Theory. Business Ethics Quarterly 18(2):15390.
Albino, V., Balice, A., and Dangelico, R.M. 2009. Environmental
Strategies and Green Product Development: An Overview on
Sustainability-Driven Companies. Business Strategy and the
Environment 18(2):8396.
Baird, N., and Rowen, S. 2010. Lean and Green: How Sustainable
Practices Are Changing Retail. http://www.rsrresearch.com/
2010/11/01/lean-and-green-how-sustainable-practices-are-changing-retail/.
Bansal, P., and Hunter, T. 2003. Strategic Explanations for the
Early Adoption of ISO 14001. Journal of Business Ethics
46:28999.
Barbour, R.S. 2001. Checklists for Improving Rigor in
Qualitative Research: A Case of the Tail Wagging the Dog?
BMJ 322(7294):111517.
Barratt, M. 2004. Understanding the Meaning of Collaboration
in the Supply Chain. Supply Chain Management 9(1):3042.
Baumgartner, R.J., and Ebner, D. 2010. Corporate Sustainability
Strategies: Sustainability Proles and Maturity Levels.
Sustainable Development 18(2):7689.
Beke, T., and Epstein, M.J. 2008. Transforming Social and
Environmental Risks into Opportunities. Strategic Finance
89(9):428.
B
oger, M. 2010. Gestaltungsans
atze und Determinanten des
Supply Chain Risk Managements. K
oln: Josef EUL Verlag.
Breckenridge, J., and Jones, D. 2009. Demystifying Theoretical
Sampling in Grounded Theory Research. Grounded Theory
Review 8(2):11326.
Carper, W.B., and Snizek, W.E. 1980. The Nature and Types of
Organizational Taxonomies: An Overview. The Academy of
Management Review 5(1):6575.
Carter, C.R., and Easton, P.L. 2011. Sustainable Supply Chain
Management: Evolution and Future Directions. International
Journal of Physical Distribution & Logistics Management 41
(1):4662.
Carter, C.R., and Ellram, L.M. 1998. Reverse Logistics: A
Review of the Literature and Framework for Future
Investigation. Journal of Business Logistics 19(1):85102.
Carter, C.R., and Jennings, M.M. 2002. Logistics Social
Responsibility: An Integrative Framework. Journal of
Business Logistics 23(1):14580.
179
180
S. Brockhaus et al.
181
182
SHORT BIOGRAPHIES
Sebastian Brockhaus (Diplom degree University of Hamburg) is a research associate at the Institute of Business Logistics
and General Management at the Hamburg University of Technology. His research focuses on sustainability in supply chain management and supply chain relationships.
Wolfgang Kersten (doctoral degree from the University of
Passau) is Professor for Logistics and Director of the Institute of
Business Logistics and General Management at the Hamburg
University of Technology. Prof. Kerstens research focuses on
the areas of supply chain risk and complexity management, sustainability in supply chain management, and project management.
S. Brockhaus et al.
His work appears in several anthologies and conference proceedings as well as other international outlets.
A. Michael Knemeyer (PhD University of Maryland at College Park) is an Associate Professor of Logistics at the Fisher
College of Business, The Ohio State University. His research
focuses on the areas of logistics outsourcing and supply chain
relationships. Dr. Knemeyers work has appeared in the Harvard
Business Review, California Management Review, the Journal of
Business Logistics, the International Journal of Logistics Management, Transportation Journal, Journal of Supply Chain Management, and International Journal of Physical Distribution and
Logistics Management.